Cohort 6 - Day 1 - 4-15-2025

I should be able to. Yeah. All right. Meeting recording in progress. Okay, that's good. It looks like it's recording, so. Yeah, it says it's recording. So we're going to go ahead and do that. And we've got Michael Espinosa that just joined. Welcome, Michael. Good to see you. Thank you. Appreciate it. Good to see you. I have my son with me for a few minutes while my wife's just finishing up her nap. Amazing. Well, hey, I had, I've had Mike, so it's funny, I have four kids. My oldest, my 12 year old was the first one he wanted to join and so a couple weeks ago he joined one of these classes and he's built a whole software business doing exactly what we're gonna be doing today. But everyone, all the way down to my 10 year old wants to do it. Right? So my 10 year old, or I guess he's 9. I'm sorry he's 9, but he turns 10 later this year. But anyway, yeah, Aaron wants to do it, Carter's already done it and Brooklyn wants to do it as well. So we like kids in the class. I think that's great. You know, if you want to bring your kids in to learn something awesome like this, I'm, I am all for it. Okay. We're just missing one person. Let me, I don't know if I have Nicole's phone number, otherwise I would text her. But let me, let me send her a quick, a quick note here and just make sure. Let's see. Nicole, right there. Tingy Glass. As started AI class started.

John, what did your 14 year old build?

So he, and by the way, he's almost 13, he's 12 still, but he built. Yeah, it's pretty cool. I'll show you. Well, I'll, I'll explain it in a second, but let me see, let me just really quick while I'm typing here, I want to get this copy. I'm going to just send her this link so she can just click. All right. Okay. So yeah, in a minute here, we're all going to build something really cool in a language called P5JS. And he loved that experience of building like a little game basically using AI that he decided to create a, a place where you can share all of the different types of games that you like. Right. That you create with this. And so kind of, kind of almost like a, an a place for anybody to be able to make a game and then publish it. Kind of like Steam almost. Right. We'll say a watered down version of Steam, which I think is pretty cool. So, anyway. Okay. All right. I don't see Nicole on here yet. And again, I don't have her phone number. I'm going to check one more. I'm going to search my email and see if I have her phone number. Right here. And if she doesn't join, then she doesn't join. But we'll send her the recording or we'll get her caught up and she might join halfway through. So we're already recording. We're going to get started. I was just saying to John Downey, he was the first one on, and I did this course this morning and we had 12 people on that one. It's very full and it's a lot of work juggling 12 people's businesses being built all at the same time with AI. And we're really going to be building a lot of stuff today. And so I'm excited for this afternoon to be a little bit more relaxed. We're still going to be going very fast. And so you're all sharp. Let me just check really quick. We're going to do a survey. Who knows the shortcut, the keyboard shortcut, for selecting everything, select all. Just blurt it out. Control A, Control A or Command A. Awesome. What about copy Control C? What about paste Control V? All right, now we don't have to go over that anymore. Control A, Control C, Ctrl B. If you can do that and switch between tabs, you're gonna be. You're gonna do just great. All right, now let's do another little thing. For those of you that have not used Zoom before, I'm sure all of you have, but make sure you know where the share button is on the share panel so you can share your screen. I'm gonna ask you to do that. I'm gonna add randomly. Be like, all right, Dustin, share your screen. Or hey, Kurt, share your screen. John, you're up. You know, and it'll be random. I'll just be kind of spot checking, say, hey, you know, or you'll say, hey, I've got a question. And I'll say, share your screen. Let's work through it together. Okay, so this is very hands on. If you have a question, I don't want you to feel dumb. There's no such thing as feeling dumb in this class. You're all learning something. We're all learning together. And if you have a question and you were the one brave enough to ask it, odds are a few other people also had the same question. So just ask it. We'll share it together, we'll work on it and we'll get it through. Okay. A little bit, a couple more, you know, just items of, or inventory items, I guess. What does everybody. Is everybody signed up for a paid version of Chat GPT? Okay. What about Grok? Okay, I'm, I'm not. Yeah, Justin, I saw you, I saw you shake your head. So you don't have to have a paid version of GROK to do this, but it will help significantly. And there is a section of the, of the class where we will want that. If you sign up for GROK on your phone, it's $50 a month. If you sign up on the computer, it's like 30. So anyway, so can I ask real quick?

I attempted to, but it's giving me an error message. Like I try and sign in. So I signed up, try to sign in and it, it just gives me an error message.

I'll tell you the truth to that, Dustin, is sign into X first on the web browser and then sign in to Brock and it'll, it'll authenticate you. It won't work the other way around.

That helps. And I wasn't signed into X, so thank you.

Yeah, and just use the sign into X option.

We'll blame Elon for that problem. So. Minions. One of his minions. I, I can't control like. We were, we were dealing with a couple of GROK problems earlier today and even a Chat GPT error that came up and I can't control that. You know, we worked through them, we figured it out, but it slowed us down maybe by five minutes just by, you know, trying to figure out what was going on with something. So the four hours. I know four hours sounds like a long time, guys, but it, it's, you know, it's gonna fly. And by the way, the, the last person who's jumping on right now, she just said. Sorry, I thought it was 3pm Anyway, so she's jumping on right now and we'll have a full, we'll have a full course here. While she's jumping on, let's just go ahead and do a little bit of intros and we'll just kind of, you know, round robin here. And I'm going to do it in what might seem like a random order to you, but I'm actually just going top left across, you know, as. As it shows up. And I think that, yeah, that is in the order that people joined. So, John, because you were on time, you get to kick it off. So here's what I want to hear, I want to just hear name where you're from and maybe very high level what your idea is for what you want to do for this course and then we'll go from there. And by the way, your time limit is 60 seconds per person.

Good to meet everybody. John Downing from northwest Arkansas. I work at the Walton Art center in Walmart Amp as a director of technology services. Also a co founder of Downeyation, a very, very, very small and very young startup in Arkansas. My idea that I hope to build out today is actually an idea from my wife. A free, somewhat small paid subscription to a bartering system online called hey Neighbor, exclamation mark, where you can trade services without having to pay a whole bunch of money for professionals.

That's so cool.

Love that.

I really love that. So I live on a, on a five acre farm and let me tell you how many things I don't know how to do. But my neighbors have been, you know, that have lived here for 40 years. They know what they're doing right? When my well goes out, okay, I need someone when I've got a big huge pipe, a main that needs to be repaired. Like, yeah, I know I can watch YouTube or I can pay someone, you know, 1400 bucks. But I know my neighbor two doors down, you know, owns, you know, a plumbing company and he knows how to do this stuff. And luckily our neighbors, you know, we have a really good just kind of bartering system. But, and this is weird, I know my neighbors better year than I did in Traverse Mountain when I live near, you know, where Dustin and Kurt live. Right. Kurt was literally my next door neighbor. He, like we were, we shared a fence, you know, many years ago. But, but anyway, and I would have loved to find out, you know, exactly what. He was my neighbor. And I couldn't have told you what Kurt did for most of the time I lived there. Honestly, like I kind of high level knew a little bit, but not really. So anyway, I love the idea, John. I think it's very applicable and, and, and really cool. Although from a naming standpoint, there is a pretty big company here in Utah called Neighbor that is, that is about. It's like Airbnb for storage. So you can store stuff in other people's things and it's pretty decent size. So just consider that in your naming conventions and marketing and confusion, things like that. But anyway, I love it. Very, very cool. Welcome Nicole. So, so excited to have you here. Where are you coming from? What, where, where are you?

I'm in Arizona.

Okay, got it. Got it. Yeah, I know sometimes the, that the Utah, Arizona mountain time zone, like you guys weird, your own weird, you know, time zone there, that changes. I never really know that's. But that's great. Doesn't change. That's the point. Yeah, it doesn't change and we all change and everyone's confused. Yes, exactly, Kurt. My wife's from Mesa and so we bounce back and forth there and you know, I just let my phone tell me what time it is. So anyway, welcome. You're not super late. We're just, we're going through intros right now, so we'll get to you last. Kurt, you're next. Go for it.

Kurt Carmack. I'm a chemist and a lawyer and I work for a startup in Houston that utilizes CO2 to make downstream products and displace the use of natural gas. That's most exciting technology I've seen in 40 plus years of looking at technologies. My idea again is my wife's idea. Hat tip to John's wife as well. This is just for me to learn this more than anything. But it's a passive income idea of having a site where you can check all the information about luggage restrictions on your flight, what size personal item, what size carry on item, how much does it cost to check a bag, what's their rate of bag loss, how do you file a claim, all those kinds of things. Just a one stop shop that would be supported by advertising and be a passive income generator.

I love it. And I've got. This is gonna be awesome. You're gonna have that built by the end of this course. I think it's so cool. Very, very.

I have four domains that I bought already.

I love it. I love it. Yeah.

Find them all in the same place and get started.

I think that's fantastic. Kurt, we'll help you build that out. Okay, so Nicole, it's going to be. When we get to you, it'll be what's your name? Where are you from? And then what's your business idea? Kind of. And what do you do right now? Just a little bit of background. And your time limit is 60 seconds. All right, Dustin, go for it.

Dustin Bitter. So live here in Utah and I, I work for USAA out of Texas. Work from home as an innovation advisor. So I help usaa, the insurance bank company, figure out what, what are the new technologies out there that we should be using. And AI is top of the list right now. So I love this kind of stuff. Personally, I'm also very into personal finance and I've Been wanting to kick off my own personal financial coaching business for a while, but the hurdles of getting started have kept me from doing that. So I'm actually really looking forward to using this to get my financial coaching startup going. Using a very specific tool that I'm excited to, to share with y' all later probably. So that's about me and why I'm here.

I love it. Dustin, did you, quick, quick question. Did you get your company to pay for this course? So.

Yeah, no, not yet, but I'm, I'm thinking retroactively at some point in the.

Near future if, I mean, if that's your job, you got to get them to pay for that. So. All right, cool.

Hunter? Yes.

My name is Hunter Vaughn. I'm from Knoxville, Tennessee. Currently I'm a solutions engineer for Epic Systems, the largest EHR in the world. So I support use of surgical software in hospital systems across the world. My idea, I have a couple ideas I'm sort of leaning against. One, I'm considering a marketplace that will allow you to search across a bunch of secondhand platforms online, read descriptions and help you find secondhand clothing items that.

Fit your descriptions across a bunch of different platforms. Very cool. I love that it's centralized secondhand. That's, that, that's really, that's really fun. So we'll be doing a lot of scraping type things with you and you know, searching cron jobs, some API keys as well. So yeah, we'll do lots of API APIs tomorrow, so don't get ahead of yourself too much. But yes, that's exactly what we'll be doing. I love it. All right, Michael. Michael Espinosa, currently out of Riverton, Utah. I'm a financial advisor. I work with 401k plans, predominantly certified financial planner. I'm here because I want to learn more about AI. So I've got a couple different ideas to kind of diversify income streams that aren't quite as stock market dependent. Number one is a kind of custom tailored to you taking a review and turning it into testimonials for kind of small and medium sized businesses to really boost their online presence. The second idea I have is a streamlined AI enhanced real time survey tool that businesses can use from everything from getting HR feedback or employee satisfaction testing markets that be subscription based and we'd help build the survey, process it, give it a dashboard so they don't have to wait weeks and weeks to get real time results back. Got it. Instead of you helping them build it, have AI help them build it. Yes, AI. When I Say we. I mean my team of AI robots. Team of AI. You'll have them enter, you know, a prompt and then it will take them through a wizard to help it all set it up. Then it'll process it all and spit out a perfect survey with a dashboard for them. You can build that, you can build all that in this with the things that we're going to learn. I love it. And last but certainly not least, Nicole.

All right. I'm actually here with my sister, she's my business partner and we do a facial head spot and, and so we're just looking to make the website a little bit better, more clear messaging and also just attract more clients and kind of streamline a lot of things. So we're not doing back and forth emailing and just kind of give a little more info up front that we're not having to do kind of all that back end stuff. But I also am a birth doula. I have a birth business and I teach birth classes and I also am a photographer. So I just am doing a lot of business startups all the time, mostly here just to learn all the AI tools and kind of keep up on all of that stuff.

Okay. I love it. Nicole, I am very much like you. I have nine things I want to do any given moment and I have so many business ideas and I literally run them. Like, for those of you that know me, you know, I'm a whitewater kayaker at a pretty high level and I do that a lot. And I'm a pianist and a musician and I put on concerts and sell sheet music. And I also run. I'm CEO of a couple companies, including the one, you know, this thing that I just started six weeks ago that it was kind of an accident and was kind of a social experiment and then it turned into something bigger. And that's fun, but, you know, and I have a five acre farm with like 10 million weeds and I have a wife and four kids and I do a ton of different things. And so obviously I've learned a lot of AI things out of necessity. Right. I've had to figure out how to, how to survive. I'm literally writing a book right now that is called Time and All Eternity. And it is really about how, you know, I want to be able to do things that would require me to have roughly 400 hours a week. And since I only have 168 hours, I have to figure out how to prioritize my life. But the consolation is that I don't need to have fomo. I don't need to worry about missing out because I am going to have a long time in eternity because I believe that that's what we're going to have.

Yeah. So, John, are you writing it or is your AI double up?

Good question. So writing is something that I'm very particular about. AI is my editor. I actually don't. I actually love the mental exercise of writing. It's enjoyable for me. And so for AI to write would be like me asking it to write a piano song for me. Like, no way, man. Like, I want to sit there and I want to write that piano music myself. So. So for all of you, you know, you're all. You're all entrepreneurs. You all have ideas. You know, I know all of you. You know, I think the person I know the least on here is probably. Probably Hunter and so. And Nicole, I kind of know you just because I feel like you already said a lot of things that are like me. And I know a couple of.

I think you're my brother at the.

Chorus last week, so I at least know your family a little bit about that. And so I don't know you. And. Nice to meet you. And Michael, you and I haven't done a ton together, but Dustin and I have spent a lot of time together working hard. We were both in a stake presidency together. And that. That, you know, that brings you close together. You know, Kurt and I have been, you know, know, neighbors and. And John and I have gone back and forth a decent amount on, you know, one of my businesses is called Velocity Access, and we're in the venue management space, and he works with a big venue. And. And so I do that and that I. I'll. Throughout the course, John, you'll get a little update on what I'm doing, you know, in relation to that and how I'm actually completely scrapping everything that we were building and completely doing it differently because of AI, Right. And what that really means and what it might mean for people in your space, too. So I think that, anyway, guys, whatever idea, during this course, you're going to pick an idea and you're going to develop it out. Some of you have a very specific idea. Some of you have, like, five ideas. One person in my class last week who's the chief experience officer for Angel Studios, for those of you that know what that is, you'll be like, oh, that's pretty cool. He literally built three complete solutions during our class because he figured it out. He was sharp and he had tons of ideas, and he just was like, I'm going, I'm building. You know, he just had multiple screens going. So. So anyway, bottom line is you guys are going to become very, very proficient with AI over the course of this next three days. These 12 hours will go by fast. I've had most. In most courses, at least one or two people stay up until 3am that first night working on their ideas because they just don't want to stop because it's really fun. Once you figure out what is possible using these tools that I've had you sign up for. How many people have signed up? Nicole, really quick, are you signed up for ChatGPT already? And what about Grok?

Yes. The only one wasn't. Was the last one. I was working on that because I thought I had an extra hour.

That's fine. That's fine. Who has not signed up for Replit yet? So just Nicole. Okay, looks like everybody else is in. That's fine. So, Nicole, go ahead and do that on the side while we're talking here. And you can do that. You can use my little code. If you use my referral code, it will try to get you to sign up for a year. But if you just click continue and then switch to monthly, you can select the monthly plan. You don't have to spend $300 right now. Replit, by the way, has doubled in price since I first. Thank you, Dustin, for letting me know that it's doubled in price since I first signed up for it. So replit's getting more popular. They're growing very, very. I had a meeting with them on Friday because they've seen some of the stuff I'm building and how much I'm using it and they're like, can we just talk to you? Can we do a user interview? And then I told them about everything that I'm doing and they're like, can we put you on our website in our marketing videos? And like, so you're going to see me a lot more around replit and hopefully I'll get some cool, you know, deals for my students and, you know, credits and things like that that we can use so that you guys can all, all benefit from that. So I am a huge Replit fan. I'll talk about why when we get there. But to kick this off, I want everyone to open. Well, actually, you're going to watch me for about four minutes. I'm going to share my screen here and then we're going to open Grok. We're going to do some fun stuff. Okay. All right. So I'm Going to share my screen. Welcome to my crazy life. By the way, I closed most of my tabs, right. And this is what's left. So that should give you an idea of kind of, you know, what, how my brain works or doesn't work. However you want to. However you want to look at that. But let me ask a question. How many of you are actual developers that know how to write code?

I'm not. I did do coding school after high school.

Oh, yeah? That's cool. And Hunter, I'm seeing you, you know.

Yeah.

I don't know what I'm doing. A little bit. Am I seeing that in your face?

Yeah.

So bottom line is, guys, I am not a developer. I don't know how to write code Now. I know how to write code and I'm actually gonna do this through X right here because I just like this interface a little better. But anyway, I've been a CEO. I've been the CEO of a software company, and I've been parts of software companies for two decades, but I've been since 2016. I founded one. I raised $13 million and did all kinds of great stuff there and, and built and sold a big software company. So I'll talk about that experience. But ultimately, all I'm trying to say is, well, I can't write code. I understand how to architect a business and how to architect a software solution. And I'm going to do my best to teach you guys that during this course, as part of what we're doing here, you're not going to become developers as a result of my teaching, because I'm not a developer. But you are going to learn how to just put a puzzle together, right? And if you're good at chess, if you're good at things like that, or you're. You're good at seeing systems kind of, you know, oh, I understand why that fits together. You're going to excel in this. If you're not, that's okay. AI is really good at filling in the gaps where you don't have, you know, that, that piece that you need. So what I want to do right now is we're going to build some software. You each are going to build your own quick piece of software in the next five minutes. And it's going to be something very, very simple. We're going to use a front end type of code called P5JS, a language called P5JS. And it's very, very easy. You all can do whatever the heck you want. For my example today, I'm Going to try to build a solar system. So here's what we're going to do a solar system simulation. So I'm going to say please write. By the way, I use please. You don't have to. But when AI takes over, I want it to remember that I was kind to it and I didn't swear at it. And anyway, that's. That's just me. Anyway, please write some code for a simulation of the solar system using P5J S. That's my total prompt for this. First thing here, we're going to use Grok. Grok is great at writing code. Okay. So is chat, GPT and certain models. But Groq 3 is particularly awesome at it. So we're going to do that. You don't need to understand what any of this stuff means. You don't like. I don't know what it means. Right. I kind of know what it means now. I know that this. Whenever you have a forward, forward slash right here, this is called markdown. And this green stuff right here is basically just explaining what it is. Right? Like, okay, here's the speed. So if I wanted to adjust the speed, I could change this. Okay, that's great. Or I could just ask Grok to do it for me. So as soon as I write this, I'm going to click copy and then I'm going to run it in this P5JS web editor by just clicking the link here. I'm going to select all, and I'm going to delete it and I'm going to paste and I'm going to click play. And hey, I've built at least my first version of a simulation. Now it's not. Wow. It is moving very slowly. You can see how this is moving, like really slowly. All right, so that's cool. Let's make it. Let's make it faster. Let's speed this up a lot so that I can, you know, actually see the planets moving. Also, let's make it more interactive. When I click, it should shoot meteor out of my mouse where I click. Let's go add. Yeah, my cursor. Whatever. It gets it. It understands what I mean. Right? Which is cool. You're exactly right. We'll talk more about that. I'm specifically, and on purpose not being careful when I'm talking here. Let's also add spinning background of stars. Yeah, whatever. Let's do that. Right. So I'm just going to say some thoughts. I'm not being careful about how I speak. It understands what I mean. It's interpreting that very quickly. And then it's going to write a new version of this code and let's see what it looks like. Okay, and we're going to copy it. Yeah. Now see how I had to wait till it was done? There's always this little copy button right here. It's going to give me instructions again how to run it. We're going to control. Do this and we'll click play. Okay, so, hey, now it's better. Now we get some stars. We're spinning a little faster if I click. Yep. A little thing is shooting out as I click there. It knew what I meant from my mouse. I can zoom in on that. And now I can keep iterating on this and make it prettier and bigger and more fun. And I'm going to. I'm going to keep working on that, but I want you all to do it now. Okay? So go ahead, go to Grok, and you're going to say, please write some code for whatever. Right. Maybe your simulation is a flower garden or your simulation is the tallest mountains in the world on a graph plotted out. Or show me a simulation of whatever you can think of. The Oregon Trail. Make a game where you have to run away from a dinosaur and if it catches you, then you die. Whatever. Tell it anything you can imagine and just say, please write some code for X. Your idea using P5J s. Okay, and you're all going to. Some of you may run into a problem or it might not give you the right type of code. And that's what we're going to debug a little bit if it doesn't work out for you. But it should write some JavaScript code and then it should tell you, go to this P5JS web editor, and that's as far as I'll go for now. So go ahead and just try it. Pull it up. I'm going to keep talking and I'm going to keep messing with my version, but in a second, I'll ask you guys to share. If it's not working or you're confused, speak up. Okay, Nicole, you might not have been here in the beginning. There's no such thing as a dumb question here. All right? So go ahead and ask it. If you're like, I don't know what I'm supposed to be doing. All right, Just copy that code, paste it in. Remember, when you first bring up the P5, the editor.P5js.org it will have some existing code. Just make sure you select that. Delete it, paste Your code. Click play.

Okay, John, if it doesn't suggest opening it up in that. That editor.

Yep.

Mine's giving me a code and then, like a preview tab.

Yeah, that's fine. So it's wanting to preview it for you in its own editor. Say. Just say, I want to use the P5JS editor. Where do I go for that? Just ask it.

That.

Make it more neon and visually exciting. So I just asked it to make it more neon and visually exciting. So I didn't give it specifics. Of course I could give it more specifics, but I want it to just be creative and do something fun and see what it comes up with. So it's going to add some pulsing, and the background stars are going to have varied hues. All right, so I'm going to copy that, paste it in again, and update it. Okay, so mine. Oh, yeah. See, this is way more fun now. This new version is way more exciting. Now, if you're still seeing my screen, maybe you're working on yours, and that's fine. My outer planets are still quite slow because their orbit is pretty significant.

So.

I'm going to ask it to speed that up. Our planets are still pretty slow. Let's speed the whole thing up so that I can see their movement. Also, add some visualization for the moons associated with each planet. Pretty cool. So you can. Who has a. Who has a visualization up and running? Anybody? I do. Who just said I do? Was it Hunter? Hunter. Yeah, Hunter, let's. I'm going to stop sharing my screen. I'll keep working on my little thing, but go ahead and share yours, and let's see what you got. Oh, whoops. Zoom doesn't have permissions. Yeah, you might have to do a permission thing or something right there. By the way, guys, if you run into an error or something, that's what. That's what we want. We want errors. That helps us. Figure something out. You figure out your share problem here, Hunter?

Sadly, not yet.

If. Yeah, sometimes, guys, it requires some permissions that require you to do, like, a browser restart. And if you have to do that, then you can leave and rejoin. And I know that's a little bit annoying, but it's better to do it now than later.

I'm on my third iteration, and that. I broke it with the third one, unfortunately.

Yeah, if you broke it and an error comes up, then all you have to do is look at what that error is, copy the error, say, grok, I just got this error. Paste it. It'll say, oh, sorry, I'll fix that for you, and it'll give you some new code.

All right. Thank you.

Yeah. Holy cow, that is way too fast. I'll share my screen while, While Hunter's still trying to figure his out. I'm going to share my screen. Look how crazy. My, my, I, I, I asked it to go faster, and, you know, it did, but, you know, now it's just like, I can't, you know, it's too fast. Right. Anyway, Hunter, let us know when you have it. Fear. Does anybody else have, have something working at this point?

I can share mine if you want.

Yeah. Let's see what you got, Dustin. While everyone else is trying to figure things out. And if you don't, it's okay, go ahead and share.

So I just a little background. Yeah, just. Oh, you know what? I gotta give permission to share.

Hold on just a second. Yeah, go ahead. Really quick. During this course, I know it's four hours. If you need to get him go to the bathroom, just, you know, that's fine, but go on me, Go on, you know, go off camera, go do your thing, come back. You don't have to announce, like, hey, I'll be right back, or whatever, or if you need to eat, which I'm going to do, because this is four hours, and I have, I'm type 1 diabetes, and so sometimes I have to drink some sugar or whatever or get some water. Like, this is a very relaxed course. Do what you need to. So go ahead, Dustin.

Yep. So just just spent my honeymoon in Hawaii on Maui and loved watching the whales migrate to the island. So I asked it to write, you know, showing the whale migration patterns through the Hawaiian islands. You can tell, like, the islands aren't a perfect match, and it didn't know what direction the whale should be going in. So I had to ask it to, in a second iteration to use, like, actual data showing where the whales actually go throughout the island.

So that's pretty cool.

Probably not perfect. Oh, and I did the click thing where I can click and a whale jumps out of the water.

Nice. I love that. I love that. So, so let me ask this. Is it accurate? Does it start in the southeast and head, you know, northwest? Like that is that they do.

But I actually had to tell it because it had him going north to south, and I said, they move south to north. Please use real data. So I don't know if it's actually pulling from real data or if it's just using what I told it.

Very cool. Well, I love that. And again, this is by the Way, guys, this is a very, very rudimentary, you know, exercise here. The idea here is to help you see, I can write code with the help of AI in minutes and then visualize what that code does. Right. That's the purpose of this, is to get everybody doing something. All right, so it looks like we had Hunter drop off for a second. He probably will jump back on after he gives permissions to his browser. Again, we want to do this also early on. All right, Kurt's going to show us what we've got going on here.

I asked it to simulate an air jet aircraft taking out from an aircraft carrier. And my first iteration, and then I had to ask it to speed up, and then I asked it to be slightly more realistic, and then I asked it to do it from the perspective of the actual aircraft carrier instead of a top down view. And so that's what you have here.

That's pretty cool. Very simple. But you can see, I can visually see that it does speed up.

Look at the code.

I mean, the original code was a lot shorter and it went through and, you know, you can see where it's filling in and scaling and showing the vertices and all that. Good stuff.

Yeah, I love it. Thank you for sharing, Kurt. Anybody else want to share? I won't require this. Yeah, go ahead, John. Let's see what you got.

You see that?

Yep.

Okay. We just hired a guy and I just, just came in my mind, like, how do I teach him about networking? Because he's supposed to be learning networking. So I was like, hey, drag two computers, Ethernet cords plugged into a switch. What I don't have working is that I want it to be showing that the two computers are talking to each other. Yeah, that's what I'm fixing now.

That's cool. Yeah, I love that. That's so cool. Everyone's going to have their own ideas, guys, and you're all going to. You know, I love the creativity of humans. Right. I wouldn't have ever thought, John, to do that. And I think that's really, really fun, being able to make it interactive in a way where you can just drag and connect and do that. And now you can say, okay, hey, when I connect, you know, show the same thing on each screen or whatever you're going to tell it to do. Right. So well done, guys. You're all developers now. Nicole, how's it coming on your end if you've been able to get it to do something cool?

Yeah, we're doing like a design, your dream room. So we're Trying to figure out how to make it like change the pieces in the room, like the bed, the lamp, like all that stuff. I was trying to add a feature where you could turn on the light on the lamp.

Nice.

Guys, remember this is just a, an exercise. We're not gonna, we're, this is like we're gonna take this and we're never gonna come back to it. Now you can come back to and do whatever you want. So my, my son, my 12 year old son during, when he took this course a few weeks ago, he, he loved this exercise so much that he said, hey, I want to build a place where people can upload their P5JS games and then share them with the world. And then people can, you know, donate, you know, a dollar to it if they love it or whatever. You know, very, very kind of low, low end stuff. And he didn't actually publish the site at the end yet. He's, he wanted to work on it more. So I can't share it with you unfortunately. But that was his ideas. He just loved this and, and he turned that, you know, this little exercise into actually a full blown business. Now what we design in this course is going to be much, much more, you know, complex. And when, you know, when we, when we launch it's going to be, you know, you're going to be able to charge someone a thousand dollars or $15 a month or whatever it is, but it's going to be much, much more complex here. So I think, does anybody else, Michael, do you want to share yours or anybody else want to share what they've got? Yeah, go ahead and share really quick. This is, this part is optional in terms of the sharing, but I don't want to, I don't want you to miss the opportunity to show off something cool that you've built. So yeah, since I had my, that's cool. Since I had my son with me, I said let's, let's track out what his, his growth projection is over the next 10 years. He's currently in the 70th percentile. His grandfather's 6, 4. So I want to see if he's gonna be taller than me or not by the time he's 13. And based off this he's not going to be. So that's great. But I had it. Yeah, I'm safe, you know.

Yeah.

At least be 16 till you can beat me up. So you know, just playing around with animating and have it get bigger with each, each size and things like that. So a couple iterations to get there. Yeah, that is cool. But now, now, now, now take this. Take a step back right now. Think about your, Your existing business or jobs you've had in the past. Maybe you're your own boss, maybe you're already an entrepreneur. Maybe you're in a company right now. But what if you needed a quick visualization of something? Yeah. You're literally in a meeting and, you know, someone says something like, I wonder what that would be like. And you can be like, watch this. And you just, you know, you just are quiet and you're sitting there building your own little thing on your phone, and you're like, hey, watch this, guys. Everybody look at this link. And you've, You've built this visualization really quick that you can share in the meeting or you can share your zoom. And all of a sudden you've got your, you know, you're a hero because you've built this awesome thing. So we lost Hunter. I'm sure he went off to figure out what was wrong with his computer, maybe did a reset, and so hopefully he comes back soon. But, but, but, but the point is, guys, you. You can realize very quickly, wow, I've built code. The first time I did this and I, I built a visualization and I copied some code that AI helped me write, and I found an editor or a visualizer somewhere where I could see it and see what it does, what it did. I felt pretty cool. I felt pretty empowered. I guess right is the right word. And there's lots of different things you can do, right? You could do. You know, I'll share my screen again here. My computer is still, you know, like going nuts here, so I should probably press stop on this, right? But, but if I wanted to, I could, you know, come into Grok and say, okay, let's make this. Or JavaScript. Let's also slow this down. Please show me where I can run some JavaScript code when you write it for me. Right. So again, I'm not being careful. I know the answers to these things and I know what I could do. But if I don't know, I can just ask and it will point me in the right direction. Right. Here's where we could do it. Let's create an HTML file or let's. Let's do it here. For simplicity, I recommend using Logo or jsfiddle. It's going to give me some HTML here and it's going to do it. It's going to do it completely differently now. It is going to use probably a similar type of result. It's going to draw the planets and the moons. This is pretty complex. Now, I've already given it a few instructions and a few iterations on it. I actually broke it one time and I said, hey, I got this error. It said, oh, sorry, let me fix this piece for you. And it fixed it and figured out what was going on.

Yeah, I fixed my crash as well while we were talking. That worked well.

Yeah. Yeah, that's great. Wonderful. So I'm going to copy that and then I'm going to come over here, I'm going to go to jsfiddle.net and I'm going to just paste it here and I'm going to run it. Now, of course, something might not work or I might be missing some css or maybe I'm supposed to put it in the HTML. Maybe I'm doing it wrong. Maybe I'm supposed to put it here. I don't know. Let's click run. Yeah, there we go. So now I did a double thing here. We'll get rid of the JavaScript, we'll go run. So now I don't know what I'm doing wrong here. And it doesn't matter. It doesn't matter that I don't know. Okay, the point here and what you're going to learn in this course is that you can start to use, and we're going to use a much, much more complex code editor that we're going to use in a minute here. But I could just screenshot this right here. I could just do on. On. At least on a Mac, it's Shift Command 4, where I can then select part of the screen and I can do that and I can drag it in here and I can say, this is what I'm getting. Help. You know, whatever. I mean, help. Or how do I make it animate? How do I make it work? You know? And it will. It will tell me, right? It'll say, okay, let me. Let me look at it. What is this guy doing? What does he mean by help? You know, it's trying to figure out. Obviously, the more specific I am, the better I'm going to get right. It looks like it's running in a browser, but there's a few issues. It's missing some elements, there's some visual issues, there's some console errors. So something's going on. Let me figure it out. Even though I wasn't specific, it's so freaking smart. Grok is smart enough to understand that while I didn't specify exactly what my problems are, it was Able to look at something, it could see the image, see what was going on, make some assumptions and then try to work through this. The meteors aren't working. You can do this. And hey, you might need to fix this. Now here's something that I'm seeing. It's giving me a bunch of random pieces here. Now here's the corrected and complete JavaScript code. Here's the HTML piece. It's still writing, by the way, while I'm talking through this. So it's doing all kinds of stuff. Here's how we're going to fix this. Here's what I'm going to say, what I'm going to try here. I'm going to copy the HTML code, I'm going to go into JSFiddle, I'm going to paste this, I'm going to click run. And hey, it looks like it's working now. It is going very fast. This is this fast animation. Now if I click on it, I'm still getting some, you know, multi colored, you know, meteors shooting out. So this is again a different thing. Now I could obviously make this bigger and I can adjust my workspace here so that I can really see what's going on. I can slow it down like, okay, we're gonna, we're gonna stop this because otherwise I might have a seizure here or one of you might. So again, we're not here to build, to build these, you know, these animations and you know, we might, right. One of you might say, hey, this animation is really cool. I'm gonna use it in my solution later on. So. Welcome back, Hunter. Did you want to share your screen now? Did you get it figured out?

I had to switch laptops, so I need to rebuild. But basically it was a simulation of a tree growing from a seed and then fruit would show up and then it would fall to the ground. So nothing monumental.

But no, but that's cool. And lots of people shared the screen. Thank you for sharing that with us. Again, for those of you that haven't ever coded anything before, this is pretty fun. If you've taken a coding boot camp, then you've seen some of this type of stuff before and that's really exciting. But, but there's so many languages out there like P5JS. A lot of you may have heard of JS or JavaScript. There's another one that I love to use Personally that's called 3js and it does a lot of 3d stuff using JavaScript and I use that a lot in my past company where I was doing a lot of 3D visualizations and augmented reality and different things like that. And by I, I mean my $400,000 a year CTO that I was paying to do all this stuff. I don't have to pay any of you guys that much. And you're all developing visualizations and code for much, much cheaper. So part of the unique. So that's kind of just the kickoff exercise that I like to do with this class is to show everybody, hey, we can all write code and then we can iterate on that code. We can fix problems, we can troubleshoot and we can work on it. And, and in a matter of just a few minutes, maybe 20 minute exercise here, we've each generated our own thing. And, and I promise you, this is a really fun thing to do during church where you're just like, I wonder what that is, right? Or I wonder if I could visualize, you know, like how many temples have been built in the last, you know, whatever, versus this time period. Whatever you want to do, right? And you could create some cool visualization and you know, show your neighbor while you're sitting there in Sunday school and be like, check this out. And. And then you never come back to it ever again. Because it's so easy to create. AI did the research about when all those temples were announced and created and launched and built and when the locations were selected. And it can do a ton of research. We didn't use the deep research functionality on grok. We will later, we'll come back to it, but it could take a moment and it could actually generate a complete table of all of that data that's really, really, you know, significant. And then, you know, able to show you, hey, this is exactly what that building schedule was like. And here's the visualization of it. And you know, let me add some cool music behind it too. So you can do so many amazing things in literally minutes by just knowing that you can ask ChatGPT or Grok or some AI to do it for you. And then if you don't know how to deploy the code or use the code or there's a problem with the code, you ask it, like, what do I put in this field? What do I put in this field? And I didn't do that with the JavaScript editor. I had just the HTML and then There was some JavaScript and some CSS and I could have said, do I need to put other things in this? And it could have said, oh yeah, let me properly break this up for you. And we can get more powerful on all of these things. My daughter made a game the other day that was super fun that basically it would display a color. It would say like the word green and then. But the color of it would be, you know, blue or pink or something like that. And then you had to select, you had to click the color that it was actually, you know, either. Either, you know, displaying or. But the color of the word was, you know, where it was, how it was actually painted. And it was quite challenging. You had to do it in a certain amount of time and see how many you could get right until you got it wrong. So she built all that. And she's 11, right? And she was doing. She was using JS fiddle and filling in and copy paste. And I watched her workflow using chat GPT and bouncing back and forth, and she was very adept at it. It's very natural for her. So in the same way that, you know, a lot of us grew up and, you know, our parents didn't grow up around computers, and then I did grow up around computers, and now our kids are growing up around, you know, AI. It's, it's, it's quite amazing. It will be quite amazing to see how technology continues to progress. So now we're going to dive into some, some more hardcore stuff. All right, And I'm just going to give a disclaimer here. I am going to do my absolute very best to slow down and let you guys catch up at every moment. And we will. And we'll have some times where, okay, take 10 minutes and work on this, everybody. And, you know, I'll get up and go to the bathroom really quick and grab a drink or whatever I'm doing. But, but if you do feel like, man, I just, I can't keep up, that's okay. This is recorded. You will be able to go back and watch it. You'll be able to catch up and, and even if there's, you know, you're mostly caught up, but you wanted to remember how to do that specific thing. Not only is there a recording here, but I have a whole folder of guides and, and step by step instructions and things like that that you'll be able to go back and forth, refer to, you know, as you, as you remember all the things that you don't remember. Right. So. All right, so what we're going to do now is I am going to just dive right into a platform here called Replit, and we're not going to have you jump into Replit yet. We're actually not going to get in here for a while, but I want you guys to just see it and maybe after you signed up you already started playing with a little bit and that's okay if you don't did, no big deal there. But we're just going to, we're going to describe an app or something we want to create and this is kind of fun. Every time I do a new course, I come up with a completely new idea and this is getting progressively more difficult because I've already built, you know, 25 things inside of Replit and I really do try to come up with something new. Okay, I haven't ever built a weather app. Let's do that. I literally just looked at my phone and I had a weather notification from Fox Weather or something and, and I'm going to build a weather app. All right. I want to build a web web based weather app that I may publish to the app stores later on. I want it to be visually unique and I want it to work on desktop and mobile. With a mobile first design, I want the, I want it to be highly interactive and want to be able to see lots of different types of layers, storms, earthquakes, temperatures, wind patterns, etc. Okay. So I'm just going to go ahead and start building and we'll see what it does. Okay. But now, obviously in order to build an app, not too long ago it would have taken, you know, some, some pretty tech savvy people that know how to code and know how to code specifically for what you're looking for. Right. Maybe you want to develop for iOS. Well, you need an iOS developer. Well, you want it to be on Android too? Well, that iOS developer doesn't know how to develop for Android. It's totally different. Right? I mean there's some people that know how to do that, but most people don't. You have to find a very specific person. So build the initial prototype. I don't care about offline functionality. Sure. Saved locations could be good. Develop an alert system. Maybe I'll do that later. Create app store ready packaging with PWA capabilities. I might do that later. For now I'm just gonna, I'm just gonna do this. And so anyway, back to this discussion. It used to require that you go and you hire people and either they have to be a co founder and so you're giving up equity or you have to pay them. And that costs a lot of money. Developers usually charge at least at a bare minimum, a junior developer is going to charge 60 to 80 grand a year. They're going to charge you 10 grand to put together a prototype, or 20 grand or 110 grand, depending on the complexity of what you're doing and what needs to happen. And then if you need a back end database, well, now you got to find a developer that knows how to do that stuff as well. In, you know, 90 seconds here, I was able to type something up and now we're starting to see the first versions, the first iterations of this app and it's building it out for me. I mean, look at this. This is crazy, right? Being able to just talk to AI with, with again, not refined language and let it do something. Now a lot of this course is going to learn how to get very specific, very refined language. That's just awesome, right? And we're going to work on that and we're going to get amazing instructions to give replit so that it knows exactly what it's doing. But it's this and it's building, it's building out really cool stuff. Now I bet none of this data is correct right now. It's using a bunch of dummy data and it's making a map now. And it's, but it's, but right now what it's doing is it's saying, okay, what do I need to make this happen? So here's a fun part of our discussion. Has anybody ever heard of chain of thought reasoning? Yeah, a couple of you have. A couple. I'm seeing a couple nods and a couple shaking. All right, so we're gonna go with Nicole. Nicole, I want you to tell us high level, don't get super detail, but how do you get ready? Tell us how you get ready in the morning. What's your, what's your routine? What do you, what do you do? What's your morning routine?

I wake up and I take off my mouth tape first of all so I can talk to my kids. Then I go in and I scrape my tongue and brush my teeth. Then usually I start getting ready for the day. Like start with my makeup routine, do that, and then get my hair done, get dressed, and then start getting my kids ready to take them to school.

Okay, awesome, Perfect. Now I want you to break down. How do you brush your teeth? Tell me how you brush teeth. Get a little more specific. Right? Pretend you're teaching your child how to brush their teeth. Right? Explain. To explain to us like that I.

Take my toothbrush off the electric charger and then I take the cap off of the toothpaste, put the toothpaste on my toothbrush, get it wet. And then I brush my teeth, and then I spit everything I just brushed. And then that's pretty much it.

All right, perfect. So you missed maybe a couple steps. You didn't close the toothpaste, you didn't put the toothpaste back, you didn't put the toothbrush back. You didn't do a bunch of things right. And I'm not trying to judge that, but what I'm really trying to focus on here is you took a task of get ready and your brain, without even remotely thinking hard, did some chain of thought reasoning and said, okay, here's a list of tasks I need to do to get ready to accomplish this main task, right? And then even in there, you can take the subtask of brushing your teeth and you can break it down into, okay, well, I got to turn the water on, I have to grab my toothbrush, I have to get it wet, I have to open the cap, I have to put toothpaste on, right? So it breaks it down to all these tiny tasks. And then you said, you know, I brush my teeth. Well, how do you brush your teeth? You start in the lower left first and then move to the, you know, here and then here and then here and then here. So you've got all these little things that are creating all these tiny subtasks. AI has gotten to a point where its chain of thought reasoning does exactly that. I give it a very high level task of build me a weather app. And I could have been, by the way, that that, you know, broad. I gave it a few more instructions, right, about kind of how I want it to act and what I want it to be like. But I could have just said, build me a weather app and it would have just assumed a lot of things, built a bunch of subtasks. Right now it's building a bunch of different things right here. You can see all the different types of files. It's building weather icons and weather, and it's using some geocoding and some layers. Search bar side panel. Building a bunch of cool stuff. I'm actually excited to play around with this and build it. This will be fun. Now I can click in replit. I can see over here a bunch of files and different things. It's done. Look at this. It's created 56 different things over here, right? All these components, 47 components here, alert card, forecast card, header, map view. And it's just considering, hey, if I need to do this, what do I need? I'm going to need some navigation. I'm going to need Some labels. I'm going to need some way to input data. I'm going to need some forms. And it's building all of that, you know, in this huge, massive project that used to take just a huge amount of resources and brain power and people and money. Right? So I raised $13 million to build a software company that I ended up selling when all was said and done for $15 million. Okay, now we made, we made a lot of money along the way. It wasn't, that was not exactly 13. And then we only made $2 million. We made probably $15 million of revenue along the way. So, hey, is that a successful business? Absolutely. Did I pay, you know, how much money did I personally invest? $30,000. Is that a win for me? Absolutely. It's great. Anyone's happy with a 15 million dollar exit now? I wasn't really happy with it because at one point it was worth 100 million and we had some bad leadership. I left the company as part of a mini exit and then the person I left in charge just was not smart. So we ended up having to sell it almost for parts when we could have sold it for $100 million. I had a 60 million dollar offer from Snapchat four years ago that we said no to. And you know, anyway, I could have made a lot more money, but that's not what this class is about. Point is, you can now build not only an mvp, you can build a fully functioning, customer ready product for less than $100. That's crazy. That is crazy talk, right? From an entrepreneurship and from a risk perspective, by the way, go talk to my therapist and ask them how hard that was for me to deal with. Raising money from 60 different VCs, family offices, angel investors, all these different things. And they hate me some days and they love me other days and everything in between, and they hate me and love me for the same decision. Right? Some people were super proud of me for not taking that $60 million offer from Snapchat and other people thought I was the dumbest person in the world for not taking it. Right. Whatever, you know, that's life. And every decision you make is going to have, you know, people are going to be judging you. But the point is, entrepreneurship is a very, very, very tough game. If you can do anything to help yourself avoid those problems and be able to prove a model without having to take on all that risk. Oh, let's do that thing. Okay. All right. Now it wants me to use a weather API key and that's great. I'm not Going to keep doing it right now. You guys got to see where it was, where it was headed. Okay. So it wouldn't be able to display any real weather information. I'm actually going to. I know I don't normally do this, but I'm going to do this because I think this is pretty cool and I want to do this. So we're going to do. Oh, of course it's going to tell me that I want. Hold on, let me see what it says here. Weather API key. So I'm just going to say that link. You know what, let's actually do this. Let's see if this is actually a real site. Weather API. It totally is. Let's find. Let's look at pricing. I'm sure there's a free version. Get started. Three day one million. Yeah, this is great. So we're going to get started here and actually go through this really quick. I'm going to make a real. Flow state. Leadership.com password. Whoops. I don't know if this password is going to work, but we'll see. I read. I sign up. Okay, we'll save that. It sent me a verification email first Activation. Okay, use. Okay, now go to the login page. Save my information. Remember me login. Now I need an API. Here's my API key. Copy done. That's all it took, guys, for me to set up an API and I continue. And now it's going to keep building and now it's using real weather data. That's how you set up APIs. In the time it took for Dustin to play and go get some water, whatever he did, he missed how to install an API. Now we'll do that again tomorrow with some more in depth things. But this is how you connect stripe payment services. This is how you connect OpenAI, which is ChatGPT. That's how you make your app into an AI enabled whatever. That's how you connect to the US Geological Survey to be able to check river data or whatever else data they have. Right. So many different companies have an API now. If we were to look right now at our preview, getting all kinds of errors here and it's doing a bunch of thinking and it's going to keep running into errors and it's going to fix the errors and you know, it's saying, hey, let's fix the implementation, right? So it's going to have some problems, it's going to try things. It's not going to work. Something's going to break, something's going to work, something's Going to break and it's going to go through a loop, and it might do this for another, you know, 15 minutes or so. Okay, so I'm gonna pause here. We've been talking really fast, and I want to just ask if there's any questions about what we're doing. We're. We're not going to go into replit yet. We're going to go and we're going to work on some business validation stuff first and we're going to put together a business plan. But any questions or comments or, you know, like, hey, my mind's blown already, type comments. Anything you guys want to share before we move on? My mind is blown already. Incredible.

How many credits did you just burn in repl it?

Oh, probably. Well, let's look. We'll see what it says here. I think when you build your first version, like the very, very first thing that it does, it's free. So I think so far I'm at zero pennies. That's pretty good. Now, had I been past the initial prototype and I'm now working on it, and I did all Those things, maybe 50 cents. Maybe 50 cents?

Seriously?

You what?

Seriously, 50 cents?

Yeah, probably about 50.

You get 25 bucks of credits every month just for the basic correct subscription price, which is a lot less than 25 bucks.

Exactly. Yep. And if you pay for an annual version, it becomes like you're only paying for your credits. Right? That's it. And if you go over. If you go over that $25 of credits, then they'll just bill you, you know, as you use it to give you. To help you feel really, you know, excited and comfortable about this. I built my entire business in three days. When I built that thing, that has now generated over $100,000 of revenue. I built the entire thing for $87.50. That was my bill. Okay? I just. I just. I can't stress enough, guys, how insane that is, right? To build my very first, very crappy version of my app Seek that I built back in 2000, 2016, the company that I raised 13 million for, the very first version cost me a hundred grand. Okay? And so we're talking a 99.9% discount here as compared to what it used to take. If you are building a business today that doesn't involve, you know, some cancer treatment or some hardcore, like, wind turbine that's going to require machining and things like that. Right. If it's software enabled and you don't have to raise money anymore, guys, you don't. You just Don't. You can build a business, validate it, get your first customers and be cash flowing in a week, in a few days. Okay, that's freaking cool. That's awesome. That brings a whole new level of entrepreneurship. Now all of you are smart. Everyone on here is smart, right? I can, I can tell. I know, I know a lot of you personally and you weren't aware that AI could do this at this level yet, right? It's going to take quite some time before the entire world knows this, but 10 years from now, it's going to be pretty well known. Okay, yeah, you can just kind of build anything even, even three years from now. A lot more people are going to know that. But right now, you guys are now sitting at the top. And by the end of this course, you will be in the top 1% of 1% of AI users. You're going to be able to say, okay, I can do anything in lightning speed. Not just using Replit, but the 15 other AI platforms that you're going to become very proficient at. But most importantly, how you start to string them together. That's where this starts to get really exciting, is when you are using inputs and outputs from ChatGPT and Grok and Replit and Claude and Mid Journey and Runway and Manas and all these different tools that are out there and saying, all right, how do we, how do we bounce from one to the other and refine the ideas and then, and then maybe even at the end, automate the whole thing so that it does it all without even your input except the very beginning and when you ask it to get your input. Right. So with that being said, you all have some business ideas that we talked about in the very beginning, some very high level ideas, some are more specific. What I want you to do now is I want you to, you're going to watch me do something really quick inside of Chat GPT and then you're all going to do it. And we're going to take maybe 10, 15 minutes where you're going to kind of work on it on your own. But this is what we're going to do here. Okay. So I'm going to share my screen again here. By the way, this is starting to, starting to work. I've got an actual app. This is cool. But I want to refine it. I want to do even, I want to do even better. Right? And this is pretty cool. This is, this is fun building, building cool, cool apps. And yeah, that's Orangeville. That's right. I wonder if I could type in like Payson. Wow. Yeah, see, there's some bad, there's some problems here. This is showing up behind the map. I can't read this. Which means the typing, the words are actually, you know, let's. This might.

The font color issue, probably.

It's a font. Yeah. So this is easy stuff that we know how to fix, but 72 and sunny. What, what is the. I want to just see. Yeah, that's exactly. This is actually accurate. The wind is 3 miles per hour. Yeah, I'm pulling the same data, guys, as my, as my iPhone app here, so that's pretty cool. I just rebuilt the weather app in one prompt and one API.

That is astonishing. All I can say is that we all need to run with this as fast as we can before everybody else figures it out.

Yeah, you know, and I agree with that. And there's, there's definitely a certain percentage of the population, Kurt, where. Yes, absolutely. There's going to be a lot of people that figure this out and start moving quickly and we're going to have some more high level discussions about, you know, well, what does it really mean to even have software anymore? Like, if everyone can build software, like, what's the point now? You know, like. And you know, those are similar discussions that people had when computers came out and Internet came out and calculators came out and factories came out in the industrial revolution. Everyone said, well, what are people going to do now that machines are doing all of our jobs? Well, yeah, obviously society advances and we change and so we will have some fun discussion around that, but let that roll around in your head for a minute. Okay, Right now we're going to take a step back and we're going to say, all right, let's do something a little, with a little bit more intention. Okay? Because what we, what I just did was just kind of like throwing stuff out there. And yeah, we got a decent result. Pretty dang good result, actually, if you ask, you know, if you ask me, but I love Chat GPT for a few reasons and I love Grok for a few different reasons, and I love Replit for a few different reasons, etc. Etc. Every platform has its own strengths and weaknesses. One of the things that ChatGPT is really good at is this projects component here. And before we dive into projects, we're going to talk about context and how AI actually understands what we want and how we can make it understand even better what we want and then how it remembers what we want when we come back tomorrow or next week or in the context of a larger, you know, a larger plan or idea. Okay. So in Chat gbt, if you're. So I'm paying for pro, maybe you're paying for plus. But if you're paying for it, Chat GPT has what's called a memory. Okay. And the memory. Sorry, I need to move this zoom thing around because it's just, you know, in my way a little bit here. Okay. So I click on Settings and I can click on personalization. And I have this memory thing here, right? And there's a little thing that's kind of out of place, but it says manage memories. Right now you can turn that off. If you don't want it to remember about you, that's fine. This is internal. It's not sharing these memories with other people. By the way, we'll just cover this right now. If you're worried about your data, improve the model for everyone right now is turned on, which means it will share whatever I talk about with some AI that will analyze it. No one's looking at passwords or things like that. It actually cleans it. It wipes that. But if you're talking about sensitive company data or you want to share something related to your own personal credit card or Social Security security number, you actually can do that. You just turn the model off, right? Say, hey, don't do that. And then it will keep everything on your model and it will be all secure behind, you know, awesome encryption. So just on that, can you do.

That with the unpaid version too, do you know? Or only the paid versions?

Only the paid version. If you are in an unpaid version, you are sharing that with. With. With open AI. Now, if you put in a credit card number or a Social Security number or something like that, yeah, it will. It will do its stuff for you. It will do what it needs to do, but it will not share that. It does scrub the data. It knows to. It's trying to not invade your privacy now. It wants to know how you talk and the types of things that you say, and it wants to understand that, but doesn't care about your personal data. It's really not trying to. It's not productive for it to know that. But anyway, you can manage your memories, right? And so if you look here, it knows some things about stuff I've talked about today. You know, it knows that I'm an entrepreneur and a composer and a consultant, and I'm actively building business. I create music and I do these things. It knows that I'm a kayaker. It knows I'm fluent Spanish and Chinese. It knows I'm a member of the church. It knows that I found it, I met these things that raised. It knows this stuff. How the heck does it know this stuff? Well, I told it. I told it this stuff, right? And I told it this stuff because I want it to understand who I am, right? And so if I say, you know, my hair is, you know, kind of a blondish brown, you know, update your memory right now, it doesn't need to know that. It's not going to affect anything unless I'm going to be talking to it about this. All right, got it locked in. Now if I go back into my memories here and I go personalization, manage memories, If I, yeah, hair is blondish brown. Fantastic. All right, that's great. Now I'm going to delete that because it really doesn't matter. I'm going to tell it to forget that. I can tell it to forget all of my memories and start over and give it a summary of who I am. Now I could also use ChatGPT and say, hey, make a summary of who I am and everything I'm about and copy that prompt and then go back into memories, delete it, and then paste that its result. And now it refreshes the memory and gets rid of a bunch of unnecessary stuff that it knows about me so I can control what it knows about me. That's one part of context is its memory. The second part is I could say, all right, here's a discussion I'm having about what's the IQ of these different AI models? And I want to know which is the smartest here and it comes here. And then I could come back to this discussion a year from now. Then I could ask it another thing. What is the best model as of today? It will remember the context of everything that I've said here. So now it's taking in consideration everything in memory and everything in this conversation. Then of course, I can ask it to consider the whole Internet if I wanted to. Now, it can't quite do that. And there's some ways to go about doing that by changing the model. But the thing that I love here is projects. So I'm going to create a new project here and I'm going to call it Weather App. I'm making a new Weather app. I'm creating a new project. GROK does not have this. Other systems don't have this. But ChatGPT has this beautiful project thing. I can add files to it related to my project. I can add instructions. I can say, please refer to Me as your highness. Act like Steve Jobs. I want to build something special. So advise me as he would advise me. Don't just tell me things that I want to hear or that you think I want to hear. If an idea can be better, tell me. If an idea is bad, tell me. Help. Actively help me improve my ideas. One thing you need to know about AI and why it sometimes hallucinates or says something that might be a little bit off is it is programmed to try and please you. It is a yes man unless you tell it to not be, right? If you tell it to not be, yes. But now it does this less today than it used to. It used to just make stuff up because it sounded right and obviously a lot of people didn't like that. But you can be very, very explicit in your models, right? And so I've told it, all right? It's going to call me your highness, which is just dumb, but whatever. Act like Steve Jobs. I want it to be a great business advisor to me in the style of Steve Jobs. Now, maybe I don't like Apple. Maybe I want it to advise me like Bill Gates. And I want to, you know, tell it. You know, I don't know how many people left Bill Gates on this call, but maybe he's. He was good at some point or, hey, I'm Give ME Bill Gates pre 2000 or, you know, whatever. And it can do that. It can do crazy stuff like that, right? And. And so you can save its context of this specific project. Now, every time you ask it something, it will refer to this, right? So I'm building a weather app. Here's my prompt. So I'm going to come in here and I'm going to tell it. Here's my idea. I could just say, I want to build a weather app. And we'll say, great, let's do that. I am using replit for the full stack development. Great, let's get started. Now I've started a new chat inside my project. It's going to respond and act just like this. Hey, your highness, this is a worthy endeavor that if we're going to build a weather app, we're not just going to make another accuweather clone. We're going to redefine what a weather app can be. All right, now it's talking like Steve Jobs, right? I mean, right? Here's how we do it. Clean, interactive, immersive, iconic, like the iPhone. Or like the iPhone reinvented the phone. This must reinvent weather. I don't know if that's exactly accurate. Because I don't think we can reinvent weather, but at least how we explore weather. Here's the vision, the name. I don't know. We got to figure something out. Strata, Ether, Nimbus. I like those. Those are pretty good. Then it's going to just start giving me some plans. Great. Because I didn't really tell it exactly what I want to do. I said I want to build an app, so now it's trying to help me do it. So now it's wanting me to kind of dive into this. So we'll let it finish its thinking and whatever and then we're going to kind of ignore it. But here's what I'm going to ask it to do instead. I'm going to say I want you. And this is, this is one of your first, like, ahas in this. In this, you know, or maybe you've already had a couple, but here's your next one. Okay. Instead of us telling the AI what to do, we're going to, we're going to flip the script and we're going to ask the AI to interview us. So I want you to interview me one question at a time about this business idea. Let's cover the name of the business, the USPs, the deal, the taglines, the messaging and everything. Oh, let's see. I'm going to actually say one more thing. The monetization strategy and everything else we need to create a solid business plan. Okay, I'm going to copy this prompt and I'm going to drop it into, into the chat. So you all have this prompt. You can copy it when you do yours. Okay? Okay. So you're going to open up a JBT in a second here and you're going to say, hey, this is what I want to do. And just type. Right? Just tell it what you want to do. Don't be, don't worry about your wording, don't worry about, you know, whether or not it sounds nice. We're going to get there. Okay, so it's now giving me. Hey, here's your first question. The name. Right? What kind of emotional response do you want? I don't know. Right. Like I liked some of your ideas. Give me five options. Maybe you have an idea. Maybe you already have a business idea. Maybe you already know the name because you have an existing business. Hey, name my company. Is this. Great. Perfect. Here's what I'm building. I'm building this feature. I'm building a new tool for this. Tell it exactly. All right, I'm honored, your highness. Based on your vision, a visually stunning, highly interactive mobile first. All right, here's some ideas. Nimbus Strata, Ether, Zephyr Trovos. Okay, I kind of like Nimbus. I'm actually going to Google that. I'm going to guess this is already there. So you know Nimbus. Yeah, Nimbus Weather. Let's see if that exists already. I don't. Okay, Nimbus WX is a spot forecasting tool. There's a little. Looks like there's something Nimbus wx, I don't know. The other thing I'm going to check is I'm going to go to GoDaddy.com and I'm going to say, all right, is this available now? Kurt's already done this. He's already got his name. He's like, I'm ready, but I'm going to look up Nimbus and I'm just going to see, hey, is this available? Search domains. Looks like nimbus.com is taken. I might bounce over here nimbus.com and say, okay, what is this.

Nimbus Clouds was available idea.

I like that. And I might go with that. Now I'm not going to actually buy this right now. Nimbus looks like an investor thing, so it has nothing to do with weather. So I'm not worried about it. It's not going to be a problem here. Yeah, Nimbus Clouds is actually a super good idea. Or actually cloud is what it would be. Nimbus Clouds.com is available. How about Nimbus.com Cloud? Looks like someone's taken it. I could check it really quickly. I'm not surprised it's taken because it is clever. It's a good idea. Now, if your idea, if you really like Nimbus. Right, and it looks like someone's taking Nimbus Cloud, that's fine. Well, what if I just go nimbus weather.com right. This still might be taken. Boom, I found it. I could do Nimbus weather, Nimbus app, getnimbus.com, whatever. There's a lot of ways you can figure out what your domain is. And by the way, this would then act as my Instagram handle, my Facebook handle, my X handle, all of those things. And if I were building a business for real, I might take the time to go and investigate that. Okay, you don't have to do this right now. You certainly can. But most of you will do this after. After this class, you'll go figure out your exact name, you'll make an adjustment or two. You'll change it inside of your app and you'll be ready to rock. But for now, I'm going to roll with Nimbus. I'll say, let's roll with Nimbus. Boom. Nimbus. And actually I'm going to be more specific and say to be more specific, my domain. Actually, you know what? I'm not even going to say that. I could say my domain will be nimbusweather.com but we're just going to roll with Nimbus. That's going to be the name of my company. Perfect choice, your highness. Clean, poetic, powerful, great. All right. What do you believe Nimbus should do better than any other weather app on the planet? So here's what I'm going to say. I haven't done a ton of research, so maybe you can help me. Where do you think the opportunity lies? The not just be another weather app. How do we really think different? Bringing in Steve Jobs there on this one. All right, so I don't even know why I want it to be different. How can I be better than the market now? Even if you think you know, I want you to ask because I can think way faster than you can and I want you to bring your thoughts into it and see what it says. So why most weather apps suck, right? Bland utility, Visual candy. OEM filters. Pre installed. Good enough. That rarely inspire loyalty. So here's some opportunity. The map is the ui. This is the interface. Make it emotional. How does it feel outside? Hey, I like that. Right? Beautiful enough to hang on the wall. Love that. Okay, so here are some ideas on how we can stand out and be different. Okay, so here's my draft concept. Great. Let's do all of it. Okay, perfect. Now it's going to go on to the next question. It's going to resume the interview. Okay. After it defines this brand. Feel locked in usp. Here's the visual style and the vibe. Right? Emotional tone, calm, curious, awe inspiring, etc. Sound subtle. Ambient cue. Yeah, I'm excited about. Or ambient cues. Not ambient. That's ambiance versus ambient. Here's some tagline options. Here's some different ways that we can work on this. Some messaging pillars when we're talking about it. So it's working through all of this stuff and it's going to continue to interview me, Right? Here's some monetization strategies. Now it's kind of gone off on its own. Normally I want it to interview me, but it's kind of gone back to a summary. So what I'm going to do is I'm going to re. I'm going to tell it. Let's get back to the interview so that you can really understand what I want and so that I can drive this vision. Okay, so now it's going to remember, oh yeah, you asked me to interview. Great. Let's get back to this interview. Round two, question three. So it picked up right where it left off. Even though it went off on a tangent, it's come back and said, okay, how do you want them to feel? All right, I want people to feel in awe because of this reason. Right. I want them to feel inspired. Okay, so I'm going to stop sharing for a second. And this is where I want you guys to now open chat GPT. Maybe you already have already start a project, tell it what you're doing, give it some instructions, right? Maybe you want it to be fun. Maybe you just want it to call you Hunter or Nicole, right? And hey, call me Nicole. This is what I'm about. Here's who I am. This is what I'm doing. Here's. And then get started on your idea. And I want you to get it started with your interview. I want you to get through a minimum of eight questions. It will also, instead of, you know, stopping the interview, it will also do 50 questions and never, ever, ever stop asking you. Okay? You don't need it to do that. Make sure it covers high level. Use my prompt if you want to have it, you know, interview you. But I want you to take the next about 10 minutes. Okay? It's 431 right now, my time. Let's go until 4:44, 41. Let's take 10 minutes. 441. And at any point during this, you know, stop, ask me a question. I am going to sit here and you know, eat a little snack while you guys do this. But let me, let me know if you need help or if you're stuck somewhere, okay? Or if you have a question about how this is supposed to work out at the end of this, just so you know, we're going to ask it to help us build a business plan, okay? So just know that that's where we're going. You don't need to get to that point yet because there's a few ways we'll do this so that it, it generates a really good output. But anyway, go ahead and do your interview, your business interview. I'm going to keep working on mine. I'm going to ask it a Few more questions while I have a little snack and get a drink here. Okay.

Hey John, real quick, quick question.

Go ahead, Dustin.

It seems like because it's an interview type style, it's super important for us to give Chad GPT the personality in order to ask the right questions. Is that true or does it just have an intuitive like generic?

It'll know now giving it. Now give it a style. I asked it to act like Steve Jobs. Maybe you have another business leader that you admire. Maybe you want it to act more, you know, spiritual because you're doing something along those lines. Church related. Maybe it's musical so you want it to think like John Williams. Whatever. Right? If definitely. If your business idea and you want it to kind of, you know, be based off of something you love, figure out who that person is. Right. Maybe you love salesforce.com well ask it who founded Salesforce. Ask the Internet. Google it. Right. Google still works too, right? Ask Google who founded Salesforce. I want you to act like this or hey, I want you to help me build something like Salesforce as we go through this. Right. There is no limitation to what you can say here. But great question, Dustin. Yeah, I eat a little protein bar here so I can get some energy.

John, how did you give it instructions within your project?

Yep, I'll share the screen. So when you build the project, it will come up with this thing, right? You've created a project.

Oh, it's right there. Add instructions. Got it.

And then you just add that instructions right there. Excellent. I'm just going through my interview. You hear me typing. That's what I'm doing. And I'm sharing my screen so you can watch if you want. So for those of you that might not be watching, my AI just asked me a great question, right? Like who is this for? And it should ask you that question if it doesn't tell it or ask it to ask you the question. Right. But I told it, I said, hey, you know this weather app is for. People love the weather. Right. I find myself exploring weather apps and trying all the advanced features and filters and I love finding tools. So this app would be the app I'm looking for. Right. So then it knows. Yes, that's it. You're not building for. The average commuter glances at the forecast. You're building it for the weather romantics, the explorers, the map tinkerer, the storm chasers, etc. Right. And then it defines my core user. Right. That's awesome. More strategic incident, more questions. Here we go. Take another bout. Three to five minutes or so here. Make sure you get through at least seven or eight questions on your interview. But it just asked me a question about interactivity and, you know, I don't really want to think about it a lot because my brain's kind of tired. So I just said, what do you think is a good approach here to really stand out? Right. That's the only reason I'm asking this. So I want to do it a little bit better here. Going to think like Apple name and I'm going to say, I agree with your plan. Let's do it. Yeah, so it's gonna, it's already on the question. Stuff it. One thing that you want to add in is you'll want to say, you know, as it starts to get into the development plan. Plan or things like that. Don't go too far into the development plan itself, but you want to tell it that you will build this initially web based and inside of replit. That's how you're going to build it. Tell it that you're going to build it web based with replit. All right, I want you guys to hone in. Let me ask, do you guys need another two or three minutes or you feel like you're getting pretty close?

I think a few minutes.

Okay, we'll give everybody a few minutes. I've been able to get to question eight or nine, but I did a few on, you know, on camera. So I'll let you guys do that. I'll be quite quiet for a minute. It. It got me to a question here where it's saying, hey, what's the very first thing you imagine a user doing the moment they open Nimbus for the first time? When you start to get to specifics like that, go ahead and just pause. Okay, I'll let you guys have another, let's say three minutes from now. So 4:46 is what we're aiming. Take another minute or so. Finish up your last couple interview questions. You can always come back and refine this later, guys, right? So everything you're doing. Don't feel like you're. You're just stuck, you know. Oh, no, like we're, you know, we're moving on and we're never going to get my idea and you can adjust it later and you can integrate those ideas later. You can keep refining this. So as you finish up your last question or two there, guys, the next thing you're going to do is very simple. Right? We're just going to, we're going to keep going and we're going to say, all right, that's enough. Please take everything we've discussed and put together a business plan and let's actually make it a 1500 word business plan. And be sure to ask me any questions that we didn't answer that you need my input on. Okay, so, so if there's something maybe that it needs for that business plan. And by the way, it's creating a business plan in a tool called Canvas. Okay. This is part of ChatGPT. You can take, you know, you can take my prompt. Here's my prompt. All right, that's enough. Please, you know, please put together, you can say use the Canvas tool if you specifically want it to do this so it actually opens up like a document and writes a document for you. And I recommend you do this that you use. You say use Canvas. Now it's going to ask me questions if it needs something, but I don't think it does. Now I don't know if we specified the monetization plan in my chat. Go to market strategy, soft launch, content marketing, community building, product growth, monetization plan Pro subscription free tier. Yeah, it's like I got this. This is a weather app. The weather app business model is well defined already. I don't need to be crazy about it. Now. I could, if I wanted to be really unique and say, hey, I really want to do a unique business plan, right? I want to, I want to do this different than other people have done it. And you might have a valid reason for that, but the reality is a lot of times you want to just follow the best thing, right? Copy and then, you know, improve upon that plan as you, you know, as you, as you grow. Right? So it says, here's a few open items we might need your input on. Right? Geographic coverage. Where should we go? What are the. Do you have any preferred data providers? Are there any community features you want, you want to white label or resell this? You know, what do you think? And so I'm going to answer these really quick. So, yes, want user accounts for the mvp. Geographic coverage for weather data should be global. I don't care. I don't have a preferred data provider. Let's keep the experience pure. Viewer for now. May add user content later. And I love the idea of white labeling this experience or I love the idea of being able to to white label this experience because it adds an additional potential revenue stream.

Hey John, what if you get a fail to generate error answer in chat instead? Does that just mean the canvas?

No, that was the ChatGPT problem. Just say Again. Say try again.

Okay.

Literally just hit try again. That just means that you ran into an error, it had a problem processing, and it'll. It won't have that problem again. Probably not. Right. Occasionally the server does something weird. So I say, after I answer a few questions, whatever, please update the plan. So when I do that, it's actually going to go back into this same canvas and it's going to edit the document that we just wrote and update it with all of the different answers I just gave it. Right. So I just kind of read through. I, you know, I took 60 seconds to answer a few questions and boom, I'm done. Right.

How did you get started on deciding what questions Open questions for completion at the end?

Yeah, yeah. I just basically said, look, you know, be sure to ask me any questions that we didn't answer that you may need my input on for this business plan. Right? Yeah. Every chat experience, by the way, guys, is different. Sometimes it asks me this, sometimes it asks me that. It's just going to kind of be random, and it's going to do it based on all of the other things that we've talked about. It interviews me in a little bit different way every time. It's kind of just a little bit random. But there is structure to it. And it does know. Right? It does know. When you get back to that chain of thought reasoning, it knows what a good business plan needs. It needs a summary. It needs a mission, an overview, a usp, a target audience. It needs core features. It needs a monetization plan. It knows those things. So if you didn't happen to cover it in your interview, because the interview went a little bit different direction and you focused more on the vibe and the colors. Right. Which, by the way, we didn't really cover in mine. Right. Well, that's fine. It's going to fill in the gaps. And so I could say, hey, you know what, let's really focus on the colors and make sure we get this right. Because right now I haven't defined it. So I'm going to ask this. I'm going to just say we didn't cover the color scheme. What do you think we should do for this? Right. And I asked it, what do you think we should do for this? Because it knows the context of the entire business plan already. Right. Okay. We do need something more than a default blue. Okay, what should we do now? It's literally writing out the colors and the hex codes that it's going to use in its design. Okay, here's the range Shadow Fog white Nimbus blue accent palette for events, alerts and data layers. Here's the different things are gonna happen. Here's a dark mode theme. Why this works great. Okay, so now we've done this. Yeah. Please add this into a branding section in the business plan and ensure we have all of this well documented for marketing purposes as well. Okay, so it's going to now update this again. It's going to add a new little section into the business plan. It's going to edit, edit the document again. And so after you've gone back and forth a couple times with it on whatever you want, and I'll let you guys take, take another three minutes here and I'll let you play with your business plan, talk back and forth to it, ask it if there's anything else, answer a couple questions and update it. Okay. Once we're done with that, then we're going to do something cool. We're going to refine this even further. We're going to move it into a new tool to get even further.

How important is having the right name right now?

You know, I love coming up with a name early on. For me, it just drives everything. It answers questions.

It's.

It's not like you can absolutely do a name last and you can come up with whatever you want it to be, Right.

But reason is I've just exhausted three of my favorite names based on competition already, so.

Fair enough. Fair enough. Well, you know, yeah, I mean, that's, that's gonna happen, right? Especially if you're doing something where other people have similar ideas or names and whatever, tell it to do it in a different language. And if you're stuck at the name, you know, don't be stuck at the name. Just say, hey, let's call it this for now and we'll change it later because that's very easy to change later on.

Okay.

Yeah.

And you can even pick one of the ones that you like. Right. And remember, if a name is taken, like if Nimbus was taken. Right. But there wasn't a Nimbus weather app, really. And so I'm gonna roll with it. Right. And if I, by the way, if I get down later and I'm actually creating this app and you see, you know, you see me post on LinkedIn, Hey, I'm launching my, you know, my first weather app, which I might actually do because I've always been a weather nerd and I've never done this before, and I'm super excited about it. And there's another one that's like Nimbus. I, you know, I might call it, you know, nimbo or I don't know, I might just change it a little bit, you know, or spell Nimbus wrong.

Hey, John, another quick question for you. When, when you're trying to answer the open questions for completion, I'm at a prompt that just says write something. And I wrote out answers, but all that did was leave me with a new prompt saying write something. I'm not quite sure how to.

That's weird. Share your screen. I want to see what it says. I want to see what your, what your UI is doing here.

Sorry, I have to.

Okay.

Here.

No rush.

Okay. You see that? Okay.

Yeah.

So right here, let me know when you're ready to dive deeper into any section or build out the next phase. And I just wrote one line, answers to each.

Oh, yeah, okay. Yeah, so, yeah, sorry. So. So you're working. Yeah, you're. You're writing in the canvas.

In the canvas. That's the problem.

Yeah, yeah, yeah, that's the problem. Yeah. So cut that out that you answered there. Yeah. So that's actually a working. It's like a word document.

Yeah, I understand. I need to write that in the chat.

Put it there. Yeah, exactly, exactly. And, and, yeah, yeah, now it'll. Yeah, perfect.

And then.

So let's see what it does here. Scroll down. So see what it. Yeah, there we go. Let's updating now it's starting to talk about the wireframes and things like that. We'll worry about the dev plan in a second. But if there's some things in here that you need, go ahead and tell it, say okay. After it gives you the answer, say okay. Please update the business plan now. Tell it to update that canvas. Sorry, let me turn off my notifications so we don't get dings like that. Guys, Sorry. Okay, guys, so you've got your business plan now next part is also. Take another minute or so. Finish up your last couple interview questions. Plan in my chat, go to market strategy, soft launch content marketing, community building product like growth monetization plan Pro subscription free tier. Yeah, it's like I got this. This is a weather app. Weather app business model is well defined already. You can always come back and refine this later, guys. Right. So everything you're doing, don't feel like you're. You're just stuck, you know. Oh no, I don't need to be crazy about it now. I could if I wanted to be really unique and say, hey, I really like we're you know, we're moving on and we're never going to get my idea and you can adjust it later and you can integrate those ideas later. You can keep refining this. We want to do a unique business plan, right? I want to, I want to do this different than other. So as you finish up your last question or two there, guys, the next thing you're going to do is very simple, right? We're just gonna, we're gonna keep going and we're gonna say, all right, people have done it. And you might have a valid reason for that, but the reality is a lot of times you want to just follow the best. That's enough. Please take everything we've discussed and put together thing, right, Copy and then, you know, improve upon that plan as you, you know, as you, as you grow, right? So a business plan and let's actually make it a 1500 word business plan. Okay. And be sure to ask me any questions. It says here's a few open items we might need your input on. Right. Questions that we didn't answer that you need my input on. Okay, so if geographic coverage, where should we go? What are the. Do you have any preferred data providers? If there's something maybe that it needs for that business plan? I mean, by the way, it's creating a business plan in a tool called Canvas. Okay, this is, are there any community features you want, you want to white label or resell this? You know, what do you think? And so I'm going to answer these rules. Part of ChatGPT you can take, you know, you can take my prompt, here's my prompt really quick. So yes, want user accounts for the mvp. All right, that's enough. Please, you know, please put together, you can say use the Canvas tool if you specifically want it to do this. So it actually opens up like a document and writes a document for you. And I recommend you do this, that you use. You say use Canvas. Now it's going to ask me questions if it needs something, but I don't think it does. Now I don't know if we specified the monetization geographic coverage for plan in my chat. Go to market strategy, soft launch, content marketing, community building, product growth, monetization plan pro subscription free tier. Yeah, it's like I got this. This is a weather app. The weather app business model is well defined already. I don't need to be crazy about it now. I could if I wanted to be really unique and say, hey, I really. Weather data should be global. I don't care. I don't have a preferred Data provider. Let's keep the. Want to do a unique business plan. I want to do this different than other experience. Pure Viewer for now. People have done it and you might have a valid reason for that. But the reality is a lot of times you want to just follow the best thing, copy and then improve upon that plan as you grow. It says, here's a few open items we might need your input on. Right. Geographic coverage. Where should we go? Do you have any preferred data providers? Are there any community features you want? You want to white label or resell this? What do you think? And so I'm going to answer these really quick. So yes. Want user accounts for the mvp. Geographic coverage for weather data should be global. I don't care. I don't have a preferred data provider. May add, you know, user content later. And. I love the idea of white labeling this experience. Experience. Pure Viewer for now. May add user content later. Or I love the idea of being able to white label this experience because it adds an additional potential revenue stream.

Hey, John, what if you get a fail to generate error answer in chat instead? Does that just mean the canvas.

That was a ChatGPT problem. Just say again. Say try again.

Okay.

Literally just hit try again. That just means that you ran into an error, it had a problem processing and it'll. It won't have that problem again. Probably not. Right. Occasionally the server does something weird. So I say after I answer a few questions or whatever, please update the plan. So when I do that, I'm not.

Quite sure how to.

That's weird. Share your screen. I want to see what it says. I want to see what your. What your UI is doing here.

Sorry, I have two screws here.

No rush.

See? Okay. You see that? Okay.

Yeah.

So right here, let me know when you're ready to dive deeper into any section or build out the next phase. And I just wrote one line answers to each.

Oh, yeah, okay. Yeah, so, yeah, sorry. So you're working. Yeah, it's actually going to go back into this same canvas and it's going to edit the document that we just wrote and update it with all of the different answers I just gave it. Right. So you're writing in the canvas.

In the canvas. That's the problem.

Yeah, yeah, that's the problem. Yeah. So cut that out that you answered there. Yeah. So that's actually a working doc. It's like a Word document.

Yeah, I understand. I need to write that in the chat.

Put it there. Yeah, exactly, exactly. And. And yeah, yeah, now it'll. Yeah, perfect. Just kind of read through. I, you know, I took 60 seconds to answer a few questions and boom, I'm done. Right.

How did you get started on deciding what questions?

And then, so let's see, let's see what it does here. Let's see. Scroll down. So see what it. Yeah, there we go. Let's updating. So, yeah, now it's starting to, like, talk about the wireframes and things like that. We'll worry about the dev plan in a second. But if there's some things in here that you need, go ahead and tell it. Say okay after it gives you the answer. Say okay. Please update the business plan now. Yeah, tell it to update that canvas.

Open questions for completion at the end.

Yeah, yeah, I just basically said, look, you know, be sure to ask me any questions that we didn't answer that you are. Let me turn off my notifications so you don't get dings like that. Sorry. There we go. Okay, guys, so you've got your business plan now next part is also not very difficult. Okay. None of this stuff is very difficult. Right. It's just a matter of this. The reason you're taking this class is because I have this just kind of journey, this workflow that's highly, highly effective and I've now taught over a hundred people how to do this and all of them come out with amazing businesses that they're super excited. I literally, the text that I just got was just like somebody from the class last week or maybe two weeks ago. Oh, brother. I'm just having the most fun with tech I've had in years. I've always been a visionary business guy with low. May need my input on for this business plan. Right? Yeah. Every otech skills. It's so liberating. I was just calling to say thank you and riff for a bit. So, like, that's awesome. That's what you're going to come out of this with. And chat experience, by the way, guys, is different. Sometimes it asks me this, sometimes it asks me that. It's just going to kind of be random and it's going to do it based on all of the other things that we've talked about. So again, this isn't highly technical. We're just going to go through the next step. So here we go. When you have this canvas thing open. And by the way, yeah, let me just cover the canvas for a second because, you know, Kurt, sometimes I assume we understand and maybe there's a question here. This canvas tool, you can actually type, right? Nimbus. I can type in it and I can work on it. I can save it, I can edit it myself and then I can ask ChatGPT to change it. Right. And so I can undo these things. Undo.

Okay. So does it use Markdown?

Does. Yeah.

Awesome.

Yeah, exactly. And so, yeah, if you copy this and paste it somewhere, it might copy the markdown. If you paste it into Google Docs, Google Docs recognizes Markdown and builds a document for you.

Right.

So, so anyway, you can, you can do that now. If I want to close out of this and get back to my chat, I just push X. I don't lose it, right. I can open my sidebar again here. It doesn't disappears my projects. If I want to get back into it, right. I can, I can ask another question and then I can come back to here and open this back up at any time. So I now have this document. Okay. So when you're in the document though, you can push copy. I can just push copy on that thing. So we're going to do that. I want you to copy your business plan. Okay.

You have to select it first or does it know to copy the whole thing?

No, it will copy the whole. It'll copy. If you're, if you have this thing open and you push this little copy button right here, it'll copy the entire business plan. Okay, Then we're going to. It interviews me in a little bit different way every time. It's kind of just a little bit random. But there is jump into Grok. Okay. And we're going to start a new chat. Okay. So open Grok back up, get to your chat and then you're going to type this. You're going to say, I want you to do some research on this business plan and provide some feedback. How can we improve it? Okay. That's your structure to it. And it does know, right? It does know. When you get back to that chain of thought reasoning, it knows what a good business plan needs. It needs a summary, it needs a mission, an overview, a usp, a target audience. It needs core features, it needs a monetization plan. It knows those things. So if you didn't happen to cover it in your interview, because it, the interview went a little bit different direction and you focused more on the vibe and the colors. Right. Which, by the way, we didn't really cover in mine. Right. Well, that's fine. It's going to fill in the gaps super easy. And then you're going to hold shift. Some of you may know this already. If you hold shift and you push return or Enter or whatever, right? Then it won't actually send it. If you just push Enter, then it's going to, you know, to get to a new line, it's going to just ask Grok. I want to do some research. And it doesn't have anything. Right? So shift. Enter, Enter, and you're going to paste your business. And so I could say, hey, you know what, let's really focus on the colors and make sure we get this right. Because right now I haven't defined it. So I'm going to ask this. I'm going to just say we didn't cover the color scheme. What do you think? Plan? Okay, Shift work on Mac as well. Yep, yep. The shift. The shift return works on Mac as well. So you're going to write this question. I want you to do some research on this business plan. Bjek, before you do that deep search, okay? Click this deep search button. Grok has some neat tools called Deep Search and think, and it even has something called Deeper search, which we won't try today because it takes longer, but it's now going to check the Internet for 100 different articles on whatever. We're going to figure out how we can really stand out here, right? So I'm going to go ahead and click Go.

Would you leave in your open questions or would you take them out?

Whatever you want. Whatever you want, right? Whatever's in the business plan, just paste it. It's going to do a lot of thinking. It's going to do a bunch of research. We're not done with that business plan. Right. And so we're going to, you know, we'll dive deeper in. But look, it's already searched 39 sources. It's been 39 sources. It's been 21 seconds. And so now it's researching weather apps, then it's enhancing the business plan, evaluating an AI integration, exploring AI benefits, is trying to figure out exactly how we're going to make this, you know, even better then. Now if I'm building a business, guys, I'm going to read all of this, every word. For the purposes of this course, we don't have time to sit here for 12 hours or we should do for this. Right? And I asked it, what do you think we should do for this? Because it knows the context of the entire business plan already. Right? Okay. We do need something more than a default blue. Okay, what should we do now? It's literally writing out the colors and the hex codes that it's going to use in its design. Okay, here's The Rain Shadow Fog white Nimbus blue accent palette for events, alerts and data layers. Here's the different things are going to happen. Here's a dark mode theme. Why this works great. Okay, so now we've done this. Yeah. Or, you know, not that it would take you 12 hours to read this, but it will take an extra. It took me, you know, 34 hours over three days to build Gen Aipi. My, my business that I built in three days. But that's because I went hardcore for three days, right. And I read tons of things and I really put thought into it and that's why I have now generated a huge amount of money with it, is because please add this into a branding section. I generated real ideas that were going to turn into money and I really put my thought into it and went back and forth and, and not only that, I wanted to understand it. Right. I care about what I'm working on and so I want to understand it. So you will want to do that at some point. For now you're going to go and in the business plan and ensure we have all of this well documented for marketing purposes as well. Okay. So it's going to boom and you're going to go back, you're going to read later. Okay, Read your business plan later. I want you to understand what's in your business plan, but we don't need to, you know, be perfect experts on it right now for the purpose of the course.

Yeah, sorry, old guy here. So in Grok, is there a way for me to save my post other than just clicking on the little save link and save the results of this, the search so I can find them later when I want to read all this.

Yeah. So your, all of your chats are saved in Grok, right? So you have, there's a history button and you can look at chat history. You can go back and look at any previous chat you've ever had, just like in chat GPT. So you'll be able to go back and find all of those, all of those things. Okay, so there's a bunch of different things here we're working on. You know, how to do this financial aspect, operational risk management, leveraging, emerging technologies, marketing, community building. There's a lot of things, a lot of examples and things that it's figured out okay, overall. Now update this again. It's going to add a new little section into the business plan. It's going to edit, edit the document again. And so after you've gone back and forth a couple times with it, on whatever you want. And I'll let you guys take, take another three minutes here and I'll let you play with your business plan.

Okay?

Talk back and forth to it, ask if there's anything else, answer a couple of questions and update it. Okay. Once we're done with that, then we're going to do something cool. We're going to refine this even further. We're going to move it into a new tool to get even further. It says it's a strong plan, but could improve a little bit in market analysis, financial projections and AI integration. This is where we think we need to do some better things here. So now we've got some feedback. Okay, Grok gave us some feedback on our Chat GPT plan by doing a bunch of research. Right. And here's all my citations.

Orton is having the right name right now.

You know, I. So I'm going to come down here to the bottom here at Grog. I'm going to push the copy button. I'm going to copy what it just did and then I'm going to bounce back in here into Chat GPT and I'm going to type or I'm going to paste right here. I'm going to say, well, first I'm going to say, please consider this research below and refine. Love coming up with a name early on for me, it just drives everything. It answers questions. It's. It's not like you can absolutely do a name last and you can come up with whatever you want it to be, right? The business plan. Okay. And I'm gonna do my shift, enter, enter thing. I'm gonna.

But reason is I've just exhausted three of my favorite names based on competition.

Paste back from Grok into here, right? All this information go. Okay, so now it's taking everything that Grok just took, you know, a minute to do and did a month's worth of research.

So.

Fair enough. Fair enough. Well, you know, yeah, I mean, that's, that's gonna happen, right? Especially if you're doing something where other people have similar ideas or names and whatever. Tell it dirch and it's gonna bring it back into here and it's gonna say, okay, let's make this even stronger.

Okay, so would you suggest then that.

If, if Groq to do it in a different language and if you're stuck at the name, you know, don't be stuck at the name. Just say, hey, let's call it.

This is still spinning. Has not produced anything that something probably went off the rails.

No.

Well, what Took a couple minutes for it to actually kick into gear. I just waited and it kept it.

For now and we'll change it later because that's very easy to change later on.

Okay?

Yeah.

And you can even pick one of the ones that you like, right? Yeah. Guys, Grog.

Second from the second you said it. I, I.

And remember, if a name is taken, like if Nimbus was taken. Right.

But did it and it's still not.

Is it just those three little dots that are spinning? Yeah. This is a UI issue, guys. Grok is having some problems and so what you're going to need to do is just refresh the page. Just try refreshing the page and see what happens. It will not lose the results. There wasn't a Nimbus weather app, really. And so I'm going to roll with it. So what it did, it will probably just start displaying the results. Right. And if I, by the way, if I get down later and I'm actually creating this app and you see, you know, you see me post on LinkedIn, Hey, I'm launching my, you know, my first weather app, which I might actually do because.

John, did you copy everything from key points below on.

I just copied the whole. I've always been a weather nerd and I've never done this before and I'm super excited about it. And there's another one that's like Scotty, the whole thing. Like it. Yeah, it can parse it. It understands that it's a response.

Well, like did you open the bubble that had everything or did you go.

Below, like, you know, I might call it, you know, nimbo or I don't know, I might just change it a little bit, you know, or spell Nimbus wrong.

Hey, John, another quick.

Oh, yeah, yeah, yeah. So good question. Yeah, so I didn't, I didn't go here and have question for you.

When you're trying to answer the open questions for completion, I'm at a prompt that just says write something.

And I wrote out answers, but really show all of the thinking and all of this stuff. That's a lot of thoughts. No, I just want the synthesis. Yes, it is. Key points and everything below. So I just come down here and I just push copy and that will copy from, from.

But all that did was leave me with the new prompt saying write something.

Key points right there. And so, yeah, if I just push this little copy button, paste it back in. Now I've refined my business plan.

I'm not quite sure how to.

That's weird. Share your screen. I want to see what it says. I want to. Want to see what your, what your UI is doing here. And with some amazing research now. All right.

Yeah. For me.

Did that work for you?

No, and sorry, I have two screws in here.

All right.

But I pasted it. I pasted it into the same thread where I had been doing the, the JavaScript programming before. So it may. Okay. You see that? Okay.

Yeah.

So right here, let me know when you're ready to dive deeper into any section or build out the next phase. And I, I don't know if that messed it up or not.

Yeah. Maybe start a new chat and ask it again. It might be better to do that because then you have new context. You have it. You know, it's going to get confused about why it has this JavaScript here and I would definitely do a new one. You want to start a new chat when you're, when you're doing something like this, because you really will confuse it. Don't do, do everything you can to not confuse it, because it will, it will do its very best to listen to you and it will, you know, very obediently say, I don't really understand why this guy wants this JavaScript thing in here, but we're gonna.

Yeah, exactly. Just wrote one line, answers to each.

Oh, yeah, okay. Yeah, so, yeah, sorry. So you're working. Yeah, you're. You're writing in the canvas.

In the canvas. That's the problem.

That's the problem. Yeah, so cut that out that you answered there. Yeah, so that's actually a working do. It's like a word document.

Yeah, I understand. That's what that was.

What. You know what I mean? Did you go ahead and do that?

So, guys, your prompt into the chat.

My prompt, what I use for that.

I need to write that in the chat.

Put it there. Yeah, exactly, exactly. And, and, yeah, yeah, now it'll. Yeah, perfect.

And then.

So let's see, let's see what it does.

Yeah, I'd love that.

Yeah, it's really easy. Right. I just want you to do some research on this business plan, provide some feedback.

That's it.

I'll paste it in the chat here. Let's see. Scroll down. So see what it. Yeah, there we go. Let's updating. So, yeah, now it's starting to, like, talk about the wireframes and things like that. Now, guys, while Kurt is catching up on that, we're going to do a second round of this. We'll worry about the dev plan in a second, but if there's some things in here that you need. Go ahead. Although it's going to be slightly different. Remember the deep search button, Kurt, right? Now if you look, tell it, say okay after it gives you the answer say okay. Please update the. Okay, Grok. Guys, there's Deep search and there's also Think. Now, if you click Think instead of deep search, it's going to do a lot more analytical thinking and it's going to process it a little bit differently. Okay, so now we're going to business plan now. Yeah, tell it to update that canvas. Sorry, let me turn off my notifications so you don't get dings like that. Guys, sorry. Copy it again. Okay, copy your. Your updated business plan. Go back into Grok. All right, Give me some. Is that I can improve this business plan. Okay, Then I'm going to paste it. I'm going to click the think button and I'm going to click Go. Okay, guys. So. Oh, and again, Grok is going to do its little spinny thing and then it should, assuming the UI is working correctly, start showing all of its thinking. Right? Thinking for six seconds, seven seconds, eight, you've got your business plan. Now next part is also seconds, nine seconds, right? So it's going to go through and not very difficult. Okay. None of this stuff is very difficult, right? It's just a matter of this. And it's going to do another analysis. Right? Hey, what do we think here? How can we do better? What. What do you think of the reason you're taking this class is because I have this just kind of journey, this workflow. That's this. How can we improve upon it? And you can go through this multiple times, guys, you could say think about it more. What else are we forgetting? What have we not talked about here? What do we need to do? You know, what else, right? So right here I'm gonna move my. Move my screen here a little bit so I can help my back. So strengthen the executive summary specifics. Right? So here's the current state. Highly, highly effective. And I've now taught over a hundred people how to do this and all of them come out with amazing businesses that they're super excited. I literally the text that I hear some improvements we can do. Here's why it matters, right? So it produces a really nice, you know, explanation here. It's giving me 10, you know, measurable or, sorry, 10 things in, you know, know last one here. Measurable milestones. And we're going to copy this whole thing here and we're going to paste it back into chat GPT and we're going to say please refine the business plan further with this. Just got. Was just like somebody from the class last week or maybe two weeks ago. Oh, brother. I'm just having the most fun with tech I've had in years. I've always been a visionary business guy with low tech skills. It's so liberating. I was just calling to say thank you and riff for a bit feedback. Paste that thing back in here and go. So the reality is, guys, you could do this, you know, 10 more times, you could, you could hone in maybe there's one specific part of your business that just really needs deep thinking and you really want it to think out of the box and you're not quite satisfied with the differentiation or you just don't love the messaging or whatever it is. Right. And go back and forth with an AI. So like, that's awesome. That's what you're going to come out of this with. And so again, this isn't highly technical. We're just going to go through the next step. So here we go. When you have this canvas thing open. And by the way, yeah, let me just cover the canvas for a second because, you know, Kurt, sometimes I assume we understand and, and you know, maybe there's a question here and bounce between two different models. Even if it's ChatGPT 4O and ChatGPT 4.5 or you know, deep research on ChatGPT or anthropic or maybe you like using some other chat, whatever, go into, go into Meta's llama models, right? If you're, you know, if you, if you, if you want to try that out. But, but bouncing it off of different data sets will get you improved results. And so even just what we did right now, the business plan right here for Nimbus, my weather app, with this executive summary, the core mission, the product overview, the usp, the market analysis, the target audience, the competitive landscape and analysis here, the core features that we're going to be building, all of these different things, like this business plan is better. You're in the top.001% of business plans. Like really, I've seen a lot. I've had a thousand people ask me to invest in their business and I have never seen a business plan this vetted. Right. Unless they actually used AI. And they did it in the last couple months because they did that really well. And I've seen a couple of them, but really most people still are building crap guys. And it's crazy because if I weren't If I wasn't trying to teach you while doing this, I could. I could be to this point in probably 21 minutes, because all I've done is, okay, I've got. I've got this idea. Ask me some questions about it. Yeah, let's go with this. Let's go with that. What do you think? Do some research really quick. What's the best way to do this? How do we stand out? Okay, great. Build a business plan. Copy back and forth, Grok back and forth.

Boom.

I can literally do it in 20 minutes. Guys, I'm here, ready to rock. And now I have something really solid that eventually we're going to feed into Replit. So that Replit has this beautiful plan and these guardrails. Most people do. What I did in the beginning is they just say, I want to build a weather app. That's their business plan that they tell Replit, and then Replit tries to figure out what you want and what you think and how it should do it. And it's not considering any of this research. It's not considering any of these things. Okay, so we've got a business plan. How many people want another minute or two to do to finish their business plan? Or you guys feeling, like, caught up?

I can't get anything out of Grok, unfortunately.

But it's still thinking. Still thinking a little bit for me for Grok. Is it still doing that spinny thing or is it.

And I've got nothing but the Canvas document.

That's okay. That's okay, Kurt. Let's, let's. I'm going to stop.

Share.

I want you to share your screen, and we're going to debug a little bit with Grok. Grok is having a hard, hard time.

It took me, like three refreshes to.

Get it to work, just to get it to stand.

All right, I'm going to refresh it again.

Yeah. How much time until you decided to refresh it? I mean, if it takes longer than 10 seconds to get past this.

Oh, I also hit the stop button in the bottom, right?

Yes, exactly. So go ahead and push that stop button down there.

Then I need to do click deep search again.

Well, go ahead and click refresh. Go and click refresh now. Now that you've clicked stop, there's your answer. Go up. Sorry, sorry. Stop again. Yes. It's just not doing what it's supposed to. Quick refresh again.

I'm a good beta tester. I break everything.

No, this is okay. And note that they put beta so that they can get Away with this. I don't know why it's doing this for you. Let's try something dramatic. Go ahead and just X out of this tab and go back to. I could just open it back up again. You. Were you on X.com or were you on.

I was on the X.com.

I'Ll do it.

Yeah, go back to X.com. okay. Like completely go back into what you were doing there. And then we'll pull up the chat. The chat history. So go back to grok.

Let me get the bar out of the way. I hate that.

Yeah, I know. Yeah. So, yeah, go ahead.

You're verified.

Yippee. Okay.

Okay. All right. So now go to history. See top right there. History.

Yeah.

And, you know, click on whatever your latest one was. Yeah. Okay, so now just say, please research this business plan and click deep search. And how I can prove it, I improve it or something like that, you know, I don't know. Tell me how I can improve it. Yeah, that's fine. Perfect. And then click go. There we go. Sometimes you just have to, you know, hit the power button. You know, I mean, basic troubleshooting. So it's going to come. It's, it's. It's going to take maybe a minute or so. So you're going to copy the results of what it tells you, put it into chat GPT and say, please refine my business plan with this information. And here's what we're going to do, actually. We'll sit here and we'll just let it. We'll let it research for a minute. When it's done, instead of copying it back in, I want you to immediately go ahead and type right now, Kurt, into the thing and say, please consider how we can improve the business plan.

Oh, in. Into grok.

Yeah, yeah. Don't press Enter yet. I just want you to type it. Yep. Please consider how we can improve the business plan. And I want you to click think. Okay. So without even having to go back into ChatGPT and integrate all the things, I want you to just have it think about the research it just did and how it can improve the business plan.

It's still doing the research too.

Yeah, there it is. So now it's done. Now it's not quite done yet. Let it type everything out. So go ahead and scroll down. Right. And again, normally, guys, you're gonna spend time, you're gonna read this, you're gonna understand what it's thinking, you're gonna understand all of its market research. Right. This isn't fake information. It just did some really hardcore research for you, and your business plan's awesome. All right, now it's done. Now go ahead and click go. All right, so now it's going to consider again, how can we improve the business plan? Hopefully we don't run into another problem here, although we might, and that's fine if we do. Now, I will say this. The more you go back and forth on this broad kind of a search, you are getting diminishing returns.

Yeah, of course.

So if you skip this step in, in what we're doing here, Kurt, you can go back and refine it later and it's not going to matter a huge, huge amount right now. It's going to matter more in some of the, the blogs and the content that we build, the website and all that stuff. You can do that. Missing it right now, Maybe you're missing 7% efficiency, but you got 93% of it, right? So go ahead and copy what it gave you. Go back to ChatGPT.

Actually, it's actually done it now.

It's.

It's starting to type it out.

Awesome, awesome. So it's working now. But bottom line is, take what it gives you. By the way, Kurt, it's going to give you a new answer. I want you to copy both answers. The one, the deep search answer and the think answer. Copy both of those, put them into your GPT and say, please refine the business plan with all of this information.

They're in the same place. Will it include. If I copy. It'll also include the.

It won't copy both. Yeah, you want to copy both of those answers separately. Copy it, paste it in, go back, copy the other one, paste it back in, and. And then say to chat gbg, please consider all of this and refine the business plan. Okay, so as you've done that, guys, you've now got a business plan. Now, the next step is not hard and it's again, not very big of a deal. We're going to close out of this, out of this business plan, and we're going to say now consider. Okay? And actually we're going to say now create a feature. No, let's say a complete feature list and development plan for this app or website or business or whatever.

Right?

And remember that we are using replit for everything. Okay? Copy. I'm going to just send this to your, to the, to the chat here so you guys can copy my thing. And I'm going to say, actually, please use Canvas to create this. That's actually the full thing that you want to use. And we will tell it to go. So now we've told ChatGPT let's create another document, another canvas, and this is going to be the development plan and a full feature list. Okay. This is what we're going to build out. All of these cool things, right? Pro only features, flow mode, passive displays, whatever. So again, you would normally go through and you would read everything, consider everything, and say, all right, do I really want that? Do I like it? Let's figure out how we're going to do this. Okay? Real time alerts, performance optimization. It's giving us a complete plan now. It's, it's phased. You know, development plan is going to, you know, think more like a normal human being and say, let's do this over the next year. We're going to do it over the next day. We're going to build the whole thing by time we go to bed tonight. You know, maybe, maybe if that's your style. But the bottom line is, guys, now we have a business plan and we have a development plan. Okay, So I can close out of this. I have a development plan right here and I have a business plan right here. Okay, I want you guys to take a minute and get your development plan ready and then it's time to jump into replit, or it's almost time. We have one more last thing we're gonna do, but I want you guys to. Yeah. Get to this point and take, you know, only take three more minutes and get to this point. So that might give Kurt a second to just catch up. Everybody else, go ahead and just get to this point and grab a drink. We've been going for almost two and a half hours here. Time is.

So how did you get to your development plan? I was probably paying too much attention to my screen.

You're gonna put the prompt in the chat.

Yep. The chat. Yeah. The prompt that I just gave you in the chat, just copy and paste that and it will open. It will create a new document in chat GPT that gives you a nice, a nice business plan here, John.

We need to clean up our development plan to say, I'm not doing this in phases or.

You can. Yeah, you totally can. That's a great question, John. Yeah. So, for example, you know, I might, I might just delete the, the timelines here, right. And just say, you know what, let's, let's not give it specific time. Let's just, let's just do this. That's, that's good enough. Dependencies, stripe for billing. I might read through Some of these things Replit, Full Stack Tools is great. Web hosting, Cron secrets, all that stuff.

Are there any red flags that we should. Obviously we're going to go through the whole thing, but what. What would you see that would make you go, yeah, no, I'm done. This isn't worth any more thought for a specific idea.

Yeah, I mean, that's something where you're just gonna have to be. It's between you and Chad. GPT. Right. Okay, let's. Let's do that just for fun. I think it's a great question. And it doesn't take long to test. Right.

So the reason I said that the first source on Grok is kind of exactly what I'm trying to build. So I was just looking at the app.

That's okay already. Let's talk about entrepreneurship for a second. Right? If someone else has your idea, should you build it? Maybe. Can you do it better? Yeah, do it better. Let's. Let's go over a quick. Who knows? Who's ever heard of the bucket list? Family? Anybody follow them? Yeah, right. Anybody know how Garrett made his money?

Snapchat.

Right? He sold an app to Snapchat. His app was called Scan. And you know what his app was? A QR code scanner. That's it. That's all it was. So he went through a coding boot camp in about 2011 or so, he went through a coding boot camp and he comes out of it. And as part of the boot, you know, boot camp, he said, okay, I'm going to build something cool, right? What am I going to do? Well, I'll do a QR code scan. He just kind of figured it. He didn't have AI to help him figure out the idea. He just decided, I'm going to build a QR code scanner, right? I'll turn off this. This thing for a second so we can see bigger or see our faces. But he basically said, look, here's my idea. I'm gonna build a QR code scanner and I'm gonna make it better, just a little bit better than everyone out there. And there were like eight. There weren't that many. But he goes in and he says, okay, I'm gonna look at each one. I'm gonna see how long does it take from the second I tap the app icon, how long does it take to load from the second I scan the QR code, how long does that take to pull up? You know, whatever do I. How many buttons do I have to click? What are the color schemes? Is it confusing? Whatever. So he goes through and he looks at every single one. He just creates some measurements. He's like, great, I've got some benchmarks. I know what I need to beat. Now. The fastest one opens in 1.1 seconds. I'm going to make it 0.7 seconds right when I scan. And I'm going to do this. I'm going to do whatever I have to do to figure out how to optimize it. So it's just a little bit better than all of the competition. And I will then, by definition, be the best QR code scanning app out there. And, like, legitimately, I'm going to have benchmarks where I can show here is why I'm better than everybody else. So he builds it out, takes them like a month. It's not that hard, right? For us, we could do it in, like, 29 minutes at this point, to build the best QR code scanning app out there. Right? But we have AI now there's another discussion. Well, so can everyone else in the world. So what does that mean? Okay, that's for us to think about. But. And. And, you know, we started to talk about that earlier, right? And it's going to really come down to, you know, what, what, what. Where is the value really created. But back to Garrett. He creates this. This awesome thing. He launches it. He starts to get some downloads, but he's like, okay, how do I. How do I get this out there? And I'm skipping ahead to day three content, by the way, because I talk about scan. I really. I've met Garrett. I've talked to him about it. But the way that he. He really blew up is he said, okay, I'm going to find the best, you know, the best writer, tech writer about this. And he identified a writer in TechCrunch. And he said, this person writes about apps that are kind of, you know, tools and things like I do. And she gets tons of viewership, and anytime she writes about an app, that app blows up. He just. He did his research and he figured this out. So he writes this writer on TechCrunch says, hey, please write about my QR code scanner. She's like, yeah, that's not really that interesting. And there's not really a story behind it. I don't really, you know, and he's like, please. And she's like, I, nah, I'm gonna pass. But he's like, no, you're gonna do this. And so he gets on a plane and he flies to San Francisco, and he shows up at her door at her Office knocks on her office door and says, hey, it's me, Garrett, with scan. And she's like, what are you doing here? He's like, I'm here to take you to lunch. I'm gonna tell you about my app and why it's so awesome and why you should write about it. She's like, no, whatever. You're here, Fine, let's go to lunch. They go to lunch, sits down, talks about it, and she's like, okay, well, now at least I have a story. You showed up at my doorstep. No one's ever done that before. So that's my angle. That's why it's a cool story now. And so she writes about it. He blows up, gets tons of downloads, it keeps going, more people write about it, and all of a sudden, he literally has 50 million users of his app, right? And Snapchat buys it for about 50 million bucks. Okay? Largely because he had tons of users, but the scanning was actually really good. It was good enough to get 50 million users. And that scanning was faster. It did some things a little bit differently. And it is still the core code of Snap codes, right? For Snapchat. Right? That functionality behind Snap codes, which is just, you know, a stylized QR code, really is all it is. So who cares if someone has your idea? I don't view that as a red flag, right? I'm building right now. John Downey knows this. I'm building a better ticketing platform. Can anyone think of any ticketing platforms out there? Because I can. There's a lot. And they all suck. They're all terrible, right? And, you know, some are better than others for sure. But, you know, I bet everyone has had a bad experience with Ticketmaster. Pain by evil people.

So you're already ahead of the game here.

Well, yeah, totally. I mean, the business model in the industry is just garbage, right? And those that are in it know all these backroom deals and things, and everyone that's trying to get a little piece of the pie, we're trying to make that a little bit better. Now, I'm going to tell you guys later, in a different part of our discussion, why my current business plan also sucks. And it's because of AI. And, well, everyone can make a better ticketing platform now, right? John can do it. John can be like, hey, I'm going to make a better ticking platform and sell it to Amp, right? And Amp, by the way, the place he works for is a big major venue owned by the Walton family, right? They're in Bentonville. So, yeah, they've got tons of money. And John can say, yeah, well, we're, I know exactly. We're making, you know, we're doing hundreds of millions of dollars of business or tens of millions of dollars of business, and I can get a little piece of this pie because I can build a better app now. Okay? And maybe he will, and that's fine. And John, don't worry, if you, if you go and build something better, I'll, I'll applaud you and say, yes. You know, I taught you how to do that and I really will be excited for you as an entrepreneur.

I can work as fast as you, John. Don't worry.

So long story short, guys, it is never a red flag that someone else is doing it now if they are doing it in exactly the same way, with exactly the same, you know, usp, the unique selling proposition that you're bringing to the table and all those things. Okay, well then maybe you need to work your, on your idea a little bit and figure out, well, how are you going to be different? Well, how can you figure that out? Talk to Grok about it, talk to ChatGPT about it. Think about it. Talk to users. Talk to the John Downey's of the world in the ticketing, right? Talk to people in that industry. Figure out why you're going to be different, how you're going to do that. The, the basic, you know, principles of entrepreneurship don't change because of AI. They're just accelerated dramatically, right? You just can come to those conclusions in minutes, maybe instead of months. And, you know, but you still want to go talk to people, right? Hey, what do you think about this new idea? You know, find someone in your target market. This is what I'm doing. Here's my concept. I'm building this. Is this something that you would buy? Yeah, that sounds amazing. Or we do this and man, it's just so painful to switch, right? Like, man, we just set up this new ticketing system and it took us like three months to get it right and it's still not right. And there's no way we're switching for years. Like, just give up, you know, I've heard that. I've heard that for sure. Okay, well, I heard some pain points in there. Switching took three months. How can I fix that? Maybe that's a new angle, something I hadn't considered before, right? So entrepreneurship is all about this iterative process of talking to the market, creating a solution, presenting it, taking feedback and then going through that loop, that feedback loop that you can do now. Oftentimes without even talking to the target market. Now, because AI knows that feedback, it knows that it's really hard to switch out a ticketing system because it's so ingrained into so many things. Okay, so my business plan has already changed significantly with Velocity Access as a result of that with my company. Okay, all right, so next step, we have created a business plan. We have created a development plan. Now it's time to start building. So here's what we're going to do. Next step, we are going to ask Chat GPT Please create. Whoops. Please create a three sentence description of this app. Simple. Okay. We wanted just a high level three sentence description. It doesn't have to be perfect, it's just going to be like this is what we want for me. Nimbus is a cinematic interactive weather app that lets users explore the atmosphere through beautifully animated real time and historical data layers like storms, wind and temperature. Designed for weather enthusiasts, it transforms forecasts into a living map of the planet, offering immersive visuals and smart insights. Powered by AI built mobile. First on Replit, Nimbus is where weather becomes wonder. Awesome. I love the description. I'm inspired by it. This might be one of those things that I actually kind of built because I think it's cool. Then we're going to come in here into back into Replit. Now I know I've already started this thing and it called it atmospherics and it named it Weather Pulse up here for whatever reason. Okay, we're going to go home, create app, whatever. You should get to this spot right here. Now I want you to paste that three sentence description into replit. Don't start building yet, okay? It's important. All right, you're going to, you're going to paste this thing, then you're going to go that shift. Enter enter thing again, right? And say please refer to the attachments for more details. Okay? And then you're going to paste your business plan and your development plan. Okay, so all you're gonna do right now, we're gonna copy them in there. We do a little bit more with it later. We're gonna copy it, Copy the business plan, go into repl it and you'll paste it. And when you paste it, it will look like an attachment right here. It's an attached asset. Okay, we'll say pasted business plan. Nimbus executive summary. Nimbus is a visually stunning, highly interactive. Whatever. Okay, there's my business plan. Then we'll go back to ChatGPT and we're going to Copy the. Oops. We're going to copy the development plan as well. Somehow it put me all the way to the top there. So here's my development plan and I'm going to copy it and I'm going to go back to replit and I'm going to paste it. Okay. So John. Yeah.

My business plan worked in Canvas just fine. My development plan. I keep getting an error from ChatGpts as Canvas is currently unavailable. Can I just drop into a PDF from ChatGPT?

Yeah, or just copy, like is it coming up in the chat?

Yeah, yeah.

Then just copy that stuff from the chat. Just copy whatever your development plan is. I, I don't know why Canvas isn't working. Again, I can't control open AI or, or you know, X AI for Grok, but sometimes they're going to have some problems. So bottom line is, what you need is your three sentence description, then your detailed business plan, your detailed development plan and then just say please refer to the attachments for more details. Okay. That is where you want to get to as soon as you get to that point. Guys, go ahead and push start building now. Don't. It's gonna, it's gonna do some thinking and it's gonna come up with another thing. Wait there, don't keep going beyond that. Okay, first, let's just start with this. So any questions?

You attach them to the prompt again, it's pasted.

Yeah, I just did. So we just took a bunch of cop or you know, text and all we're doing is paste and it will create this little attachment.

Okay, great.

Yep.

Good question. Oh, excellent. Okay, so now you've got. So now what we're doing, remember most people and if you go on to like Reddit and you or Google and you say is replit any good? You know you're going to get some people that say this is amazing. And other people will be like, this thing sucks. Like it has no idea. Right? Well, the people that say it sucks probably just went in and said, make me a weather app. And they're a developer and you know, or a non developer, but they're going in, they're just giving it so little information or context and expecting it to just build something perfect. Okay, you said hit start, right? Yeah, go ahead and hit start. Yep. So I'm going to go ahead and do it on my screen now, just waiting for people to catch up if they wanted to see this. But now it's configuring the app, it's figuring it out. All right. And now it started A new project. And even though I started that one earlier, I'm just. I'm starting over from scratch because it's so fast. It's going to catch up here and it's going to do something way better than what I gave it the first time. Okay, so I'm going to analyze what you're trying to build based on the information provided and create a plan for your Nimbus weather application. Okay, so now it's going to build the initial database. Now it is saying it's going to use REPL dot support for PostgreSQL database. Why? Because I asked it to. In my development plan, I told it to use internal things and replit auth. I told it to use all these internal tools. Now you're going to come up. One of the options might be create a PostgreSQL database. One of these things you can check as an additional option. If that is check it. We want a PostgreSQL database for what you're doing. We'll talk more about why that matters as soon as we get this going. In a second, we're going to do some. We're going to take you guys all through a coding boot camp in about 10 minutes. And the bootcamp really is going to be just to help you understand a little bit of language around coding so that you're smarter in replit. So would you like any of these additional features? AI enhanced forecasting using graph cast like models. I don't know what that means, but it sounds pretty neat, right? Real time alert for rare weather events. Yes, I do want that. That was something I was talking about. Flow mode for ambient display on tablets and home screens. It's pretty interesting. It's pretty cool. And I'm going to click it just so it remembers it. It's not going to build all of it right now. And then advanced layer customization analysis tools. I don't know what that means, but I like all of these things. So, you guys, this is just a choice. You pick what you want, what you like. You can click it all. It's not going to build all of it right now. Right now it's just going to kind of build the initial prototype and then it's going to go. And then it gives you a timeline. Two minutes. Within two minutes you'll have at least a preview like we saw in the beginning. And within 10 minutes, you're going to have your first app version. Any questions about this? Once you're ready, go ahead and click Approve Plan and start.

If we didn't tell it in the beginning what sort of authorization or authentication to use. Mine came up with basic username password authentication. Should I just leave it there? I don't know how to edit that.

Yeah, so if you want to edit the plan, you can just tell the agent. You can say, let's make sure we use replit's internal database and capabilities, and that will be a blanket statement for everything. Okay. And then. And then just click go, and then it will come up with a new plan for you to approve. Well, look at my cool clouds Nimbus. I don't know what it's making right now. It's doing something crazy on my screen.

That's cool. Love that background.

Even create a little cloud icon for me here. That's fun. So once we get going now, once you're going, it's going to take like 10 minutes. And you can sit there and you can watch your, you know, you can watch what it's building. Okay. In the meantime, we're gonna. We're gonna chat a little bit about. About software coding Quick Start guide. So you guys can just kind of sit back and listen for this part. Again, you're probably going to be sitting there watching, you know, maybe a mix of, you know, my screen and Slash, your awesome new app that's being built. Because it's really exciting that your app is now finally starting to get some life, by the way, as it builds it out. It's not. Not everything's going to be perfect. Not everything will work. The buttons might just be placeholders. Don't worry about that. We're going to get in and develop more. I'll show you how that works later, how we start to make things come to life. Right now, it's just building what it thinks you want. Okay. And it might get a lot of it wrong, and that's okay. All right, number one, we could tell it to try again, or we could say, hey, you know what? Let's completely rethink this. Let's, you know, scratch that. Let's do this section again, right? So we'll worry about that stuff later. For now, just let it. Let it do that. Let's learn about coding. Okay, so some of this for, for many of you might be very, very basic, and that's okay. Sit back and ignore me. Or, you know, keep working on your app. Keep looking at it, but at a very high level. It's. It's much easier to use real, real quick.

John.

Yeah, yeah, go ahead, Michael. Oh, my. I got a page unresponsive. A Couple of times. Should I exit and refresh it on replit or should I go ahead and just hit. Yeah, just wait. Just click wait. I've seen that that happens. If you're on maybe a little bit older computer, you don't have a computer that has a ton of ram, a ton of memory, then it can do that. My computer has 64 gigs of RAM, maybe 128. I think this one's a 64. I don't remember what this is. Yeah, this is a 64. So I have a lot of, you know, use capabilities. It uses a lot during this, during this section. And even while it's going, if you just click wait, it'll. It'll figure it out. If you are going to do a lot of hardcore development, it's better to have a better computer. It really is. Now I will say this is an M1. Gotcha. This computer is not like super new. This is an M1 Max 20, 21. It's four years old. But I gave it a lot of memory, right? I maxed it out. This was a $6,000 computer when I bought it. If you were to buy a new MacBook Pro, that's, you know, 32 gigs of RAM, but it's got an M4 chip in there or something like that, like it, it's going to do great, right? But if you're working on something that's eight years old and it only eight gigs of ram. Yeah, you're going to struggle a little bit. That's just part of life. All right, so front end versus back again, guys. Some of this might be very, very basic, but if you understand this language, if you understand what you're seeing and how to talk, you're going to save yourself a lot of time trying to figure out what to say. Like, for example, maybe, or maybe you haven't heard the term or this little ui, which means user interface, which means what you're looking at, right? So you could say to replit, hey, the UI feels weird here. Or you could say, you know, these buttons are kind of weird and I don't really like the colors. And you know, great. Like you could take a long time explaining what UI means or you just say ui. So this is why this is going to help you. Front end versus back end. Front end is what you can see. Back end is what you can't see and what's working behind the scenes. So front end is going to be something like a form, something you type into, something you click the way that something looks and all of that works together to form a user interface. And then that user interface and everything that you click and touch and interact with is going to interact with things behind the scenes, right? So for example, you might have a login form where they enter their email and password and then the back end is going to process that, right? And the languages you might see in the front end, HTML, CSS, JavaScript, React Vue. This is not an exhaustive list. These are just a few examples of things. So, you know, some of you might know that CSS stands for basically how things are formatted, right? Colors, fonts, sizes of text, things like that, right? Those things are going to be determined by this. And so if you know, hey, I think the CSS is messed up here, right? Or you could say, I don't like the font, either one of those work. But you might just get a little, you know, you might get farther, a little faster if you have this information now. Don't feel like you have to memorize all this. You will get this document at the end of this class. You're going to get access to this entire folder. And there's all kinds of documents for the software, all kinds of documents for overviews and different things like that, step by step. Here's a financial projection type thing. Here's some advanced workflows that I'm actually still working on. But as I continue to add things to this doc or to this folder, you're going to get, you know, you're going to get more and more stuff. And that actually bothers me that that's an untitled document. If I click here, it won't be untitled anymore. That is now fixed. Okay, so anyway, back to coding. Back end is going to be where we're going to process the data and handle requests from the front end, right? So if you try to log in, when you click that button, it's going to send the email and password check a database. Hey, does this email and password combination exist? Yes. Great. Well, let's send a response back until the front end and then the front end will say, login successful or login unsuccessful or user combination does not exist, or user does not exist. You can have it do whatever you want based on whatever you want to tell it to do. Now, the way that repl. It works is you just get to be the user, the QA guy, right? You're just going to say, hey, you know what? When I click on this button, it does nothing or it doesn't do what I think what it's what it's supposed to do. It says animate, but I click it and the app just crashes. Oh, hey, let me fix that. And it will go in, it will figure out what's wrong. Or hey, when I clicked search location and I clicked enter, nothing happened. Oh, well, let me fix that. Right? So you just have to talk to it and say, hey, this is what I did, this is what happened, this is what I want to happen. Right? And so you don't have to be an expert in order to be able to talk to repl it. But the better you understand this language, the faster you'll be able to get through it. So some common Python flask, Django, Node JS Express, postgresql. We've already brought up this word, postgresql. It's related to databases. Firebase is another thing that can handle a lot of backend stuff. Right now Firebase is an external system that is run through Amazon, right? And you can use, so you can use that or sorry, it's run through Google. You can use aws, you can use Google, you can use third party services out, you can use Microsoft Azure, you can connect to these other systems. Even using replit, it can build with all of these different things. You just have to tell it and we'll use APIs and keys to make that happen. Okay, but the backend is going to process data. It's going to handle authentication security, it's going to communicate with databases, which is where all your data is stored, and it's going to manage other types of business logic. So again, it'll check a user's login credentials against the database and return a response. Right. Or an authentication token if it's successful. A database is like an Excel document. It's a database where a bunch of information is stored. You're going to have lines or rows and columns of information and it's going to talk about how that in the database and the back end will determine how the relationships between all of that data operates. Right? And you've got a bunch of different ways that that can happen. We're going to use inside of Replit, a PostgreSQL database. If you want to use a different type of database, you can and you can connect it to replit via an API key. The the database that's inside of ReKit is a PostgreSQL database and it is the right way to do things for most use cases. Okay. If you're a developer, you might have opinions about this and that's fantastic. Do what you love. There is no rule Here, right. Replit is trying to do things in the most efficient way it knows and with the tools that it provides internally. So inside of Repwit, you have this. You have these kind of working. You have, you know, you're talking over here and you're previewing over here. You're seeing the progress tab. You can open up other things here. There's database, there's web views, there's networking, there's deployments. We'll go over all these things so you don't need to know it all. By the way, I don't even use half of these. I use webview. I have one tab with my progress that's showing what's happening and I can see a preview of it at the same time. Or I can just go to my web view, I can go to my database here and I can see my data. You might click on this database thing and it will say, hey, we don't have a database set up. Let's set one up. Okay, great. Well, it's created a PostgreSQL database. It's just barely done that over here. It's also built a ton of other things like we were showing earlier. Okay, so we'll learn how to use a little bit more of this later. So right now, a database, basically it stores data, right? Stores data in rows and columns. It's going to have unique identifiers for records and it's going to show relationships, right? Maybe we have a user and we want to know what are the users preferences, what things has it liked for my thing? What are its saved cities? I have Payson, I have Sandy, I have Alta Ski Resort, And I have St. George, and I have Mesa, Arizona. And those are my cities, right? And so I'm going to have a relationship between the John user and cities, which is another table. Here's a table, right, with, you know, here's the city, here's the zip code, here's its geographic coordinates for gps, you know, whatever. So you're going to have a bunch of different information and you don't have to design all of that information and all of those relationships. But it's really quite helpful as you, as you grow inside of this to be able to go into your database and understand what you're looking at, right? Why do these tables exist? Right? I've got a locations table. Now it's real. Now we're talking about real data. I've got a notifications table, I have a sessions table. I have users, I have weather preferences. I don't know what These mean yet. But at some point I'm going to come in, I'm going to look at what are these things? Temperature unit. Like maybe my weather preference is I want to see things in Fahrenheit, right. That's my default. I want to see things in miles per hour wind speed unit instead of kilometers per hour. Right. And so it's already starting to think through those things. So right now if I go to my web view, it's saying, hey, you know, we've got an error here. I don't know why. Right. So it's currently implementing some premium features. Got a problem here. And so I'm actually going to do this really quick. I'm just going to say I'm going to screenshot my error and drop it right here. Same. I'm seeing this error and it'll read the error, it'll look at, it'll figure it out, it's going to go work on it.

Okay.

And while it's working on that, I'm going to get back to teaching you guys.

Okay.

Now, a database. Yeah, we're going to, we're going to build some databases. That's cool. Now there's a thing called an API. An API basically connects. The best way to explain what an API is, it's connector. It's going to connect one thing to another. A button to a database, a database to a database, a service to a database, to a response, whatever. You don't have to worry about all those things. Anytime an API is needed, it will either create one or it will ask for one. Okay, so maybe there's an API, a connector, to connect your login button to your database. You won't have to tell it that, but maybe you go and you type in, you know, you go register. It's created a registration flow and you create a password. You type in, you know, Kurt Carmack and your password is admin. Right. For testing purposes and all these things, right. And you click create new user. Nothing happens. What the heck, right? I thought it was making this registration flow. Well, maybe it just hasn't connected it yet. So you just say to repl it, Hey, I just clicked or I filled out the registration form, I clicked register. Nothing happened. Oh, okay. Replit will know. Well, I haven't connected the API yet to the database. Let me do that for you. Okay. All right. So that's an API is just going to connect something to something. Maybe I need weather data, which we did very early on, right? It said, hey, do this. And I went into a weather something. Weather API and I grabbed a little API and I copied and pasted it into replit said, boom, let's add this in. Okay, so now it's still working on stuff. It's working on some map display errors and things like that. It's going to try to figure out what the heck's going on. I see the server connections reset. Let's update this and see if we can make this actually work. Okay, so it's still working. It's still trying to figure out what's going on. It's still having an error. I don't know why. Right. It's trying to create some sort of map visualization. I don't know why it's doing that, but it's working on it. It's going to keep resetting, it's going to keep trying. Let's try to make it look better. Let's try to work on this and make this better. And I can say, you know, it's still trying to have this problem. So I'm just going to keep telling it, it's having this error, it's having this error. Eventually it will solve the error and we'll be moving on. Okay, API endpoints. An endpoint is a specific URL that points at a service or some data. And a lot of times, like for example, you might have a Stripe endpoint and it will go to that endpoint, but it's going to say, hey, I need an authentication. I need to know which account I'm trying to pay. Right? Because if someone puts their credit card information does it, it needs to know where it's going to send that. So you don't need to understand all of this because replit will tell you, hey, I need an API key. Go here, then go here, then go here and copy and paste this from your account, just like you saw me do really early on in this, where I grabbed an API key. We'll work on that more later, we'll do that more tomorrow and we'll try it a couple times. And right now it's failing a bunch of times. It's trying a bunch of things. It's starting it. Actually, no, right now it's waiting for something. So it needs a session secret. So we need to set a session secret environment variable for securing the user authentication sessions. Right. Please provide a secure random string for this purpose. Now, what if I don't know what kind of session secret to create? Okay, now I know I could just say, you know, tacos right now, I've told you guys my session secret. So now it's not secret anymore. So if you guys wanted to hack my weather app later on and you knew that my session secret was tacos, you know that, right? Maybe I don't know that. Right. So what I could say is, what should I use for my secret? I'm not sure. Okay. And I could do that and it would tell me. Okay, well, you could use this and it would give you a random string, or you could say, hey, do this for real. I'm. I'm not going to use Dockers, because I think that's a great idea. So I'll just type in something else that's a little bit longer. This is just a random string, by the way. The value doesn't even matter. I could type asdf, you know, whatever, and continue. Okay, so there's a couple ways I can approach that, but it's going to ask me what do I want to do. Continue. All right, great. I'll use this. Now I'm going to add this to my account secrets so I can access it later. I can go back and look at that. There's a section, a tab here where I can go to Secrets, right here. And it's going to save all my API keys. Right? Here's all of my account secrets for my session secrets, my app secrets. I have account secrets as well, by the way. By the way. Right. It's got things like my app URL, it's got my weather API key, so it might use that later on. Maybe a stripe key. Right. Maybe a send grid key and OpenAI key. This is things that it's used in my other platforms. Right. My other apps. Right. And so I don't have to create a new one every time. I can just say, oh, hey, that's my key that I like to use for this app. We've put a few in here. Okay, so it fixed the error. Let's see if we're actually running here. Let's quick refresh and see if it actually has fixed it. So it says it fixed it, but I don't know if it has. So, like this say, still not working. So you can see I'm just kind of in this. In this constant state of debugging almost and doing QA and ux. I'm taking the user through the experience all, you know, until it works. Right. We'll. We'll talk more about that in a minute. I'm just going to let it. I want it to keep going while we, While we go through this. I want it to be working while we learn. Okay? Now an API is going to do a few different things, right? Here are a couple key concepts. This is pretty important when we're doing something. Almost everything in every app that you've ever used is going to end up either getting, posting, putting or deleting. There's also another phrase for this is called crud. And we'll come back to that in a minute. Get retrieves data. Okay? If I'm playing a video game and I click, you know, I go to a store, I'm playing World of Warcraft and I walk up to a villager and I open his shop, okay, when I open the shop, it's going to retrieve a bunch of data. Well, what's available in the database at this shop, right? And there's going to be some pre programmed, oh, I can buy, you know, a pickaxe and I can buy, you know, a cloak. I can buy whatever, right? And so I can program or pre program or have it randomized, whatever. But it's going to retrieve some data, it's going to display it. Okay, then maybe I need to post, I need to send some new data, right? Maybe I need to buy something. So I'm going to click, you know, purchase for three gold, right? This pickaxe, right? And it's going to post, it's going to send some data and say, hey, the user is sending some gold to the database and the database says that the price is 3 gold. And so it's going to do this, it's going to create something. Something, right? So it's gonna, it's gonna retrieve data, it's gonna send some new data or it's gonna update some existing data. So hey, the user had an old battered cape and now it's going to update it with a new, nicer cape that's going to give it these variables on its character. You don't have to be able to think through all this, but just understand that every time you're doing anything, whether it's in a game, on a website, on whatever, it's really doing one of these four things. It's retrieving, it's sending, it's updating, right? Hey, I forgot my password. Well, let's have a put function. That's that updates the password, okay? Then we're going to remove the data, okay? Delete, get, post, put, delete, crud, create, remove, update, delete, right? So those are the different things you can do. Or sorry. Create, retrieve. Yeah, create, retrieve, update, delete. That's what it is, okay? That's what crud is so let's say you decide to build Genaipi, which is what I build. This is the company that I built in a weekend. This really complex thing. And I decided I wanted a blog. Let's say you decide you want a blog. I built this beautiful, nice system here and I said, okay, I need CRUD functionality for my blogs. I need to be able to create a blog. I need to be able to delete a blog. I need to be able to edit a blog, update a blog. Here it is. This is the front end where I'm just viewing it. It's been viewed 26 times. I can share it on Twitter, right? This is awesome. I don't know how to create this. I just said create a share functionality. Include Facebook, Twitter and LinkedIn. By the way, I actually didn't say that we need to update Twitter to X. This is out of date even though I created this like three days ago, right. So I need to go in and say, hey, Twitter doesn't exist anymore, but X does. Please update this. Right? And the way that I would do that is I would go into here and I would tell it to do that. Okay. So what it's saying here is that it's trying to do some things. It's not gonna, it's not able to do it. So it needs an API. Luckily I have another system that uses MapBox. Oh, you know what? But that's in a. That's in a totally separate account, so that's okay. Anyway, I could go create a mapbox account. I might have to do that. I created another app called echoz using. Using replit. Right? This allows you to share. It's like a visual pin board everywhere I visited in the world, right? So I've been hearing it here and here. This is actually fake data. So don't, you know, but, but I can, you know, this fully functioning map, I can zoom in on it. I can. This I can click on. Hey, this is Bridal Veil. View my echo. This is actually information from. So this is all fake data, right, That I uploaded. Remember this is me at Yellowstone. But anyway, bottom line is I created a map based thing. So I've done that past. So now I know how to do that. I can go into mapbox, I can grab my API and update it. Okay, So I might have to go do that in a minute and make that happen. We'll skip it for now. Okay, so let's bounce back in here and finish this up. So we've gone over front and back end databases is where we store data APIs is how we talk to different systems, connect different systems or connect a button to a system, something like that. Authentication is security login, right? Authentication means login. Oauth means you're going to use a different type of system. Maybe we're going to log in with Google, log in with Facebook again. You wouldn't need to know what OAuth is other than the fact that if you say, hey, let's create a Google login, it will replit will say, great, I'll create an OAuth system. And you would then know, okay, yeah, I know what's going on now. So you could learn that by just talking to Replit. But you can say, hey, I want to, I want to be able to log in with Facebook. Great, I'll do that. Hey, I want to be able to log in with, with an authentication where it texts me, right? Where it texts me a four digit code. Just tell it, right? And it will help you understand what that is. Okay. And we'll talk, it will walk you through it and that will use a Twilio API or something like that. Okay. AI agents hosting and deployment. At some point we're going to deploy this app. Now, Replit, one of the reasons why I like Replit. Let's do this for a second. Lovable.dev. does this look familiar to you guys? Idea to app in seconds, right? Ask lovable. Hey, you know, make a weather app, right? And I could take all those same prompts and everything that we've done, right? And I can go right now. It's going to want me to sign up for free. Sign in. It's going to look at the pricing, all those different things, right? There's lovable, there's cursor. AI. Cursor is also another. It's not cursor. AI, it's cursor. Yeah, there we go. AI code editor, right? Hey, create something for me, right? What's my pricing like? Hey, I want to create a quick app, right? There's a bunch of competitors out there. Not a bunch, there's a few, like good ones that are competitors to replit. The reason why I love Replit is because it is so complete. The solution is amazing. I can do everything inside of here. I can run my database, I can run all of these different things. And this is, you know, by the way, we've spent so far only 75 cents. I've spent 75 cents today on Replit. Pretty awesome. Okay. But I'm going to click on this usage button and I'm going to show exactly how this works. Now you're going to see a big number. I've spent maybe a couple hundred bucks this month or maybe 100 bucks this month on something, or maybe I haven't even used it because I have so many credits because people keep signing up with my thing. I spent $136 of my additional credits. I've spent $71 this month on Agent. And then look at all these things that it does for those of you that are developers that have ever built anything. Autoscale deployment. We'll talk about that later. PostgreSQL storage, object storage. There's all these different things. It all can be done inside of Replit. Now, Replit doesn't build its own database system. It uses Neon. Neon tech. Okay, so this is a postgres SQL database system that's out there. The way that Replit makes its money is it, you know, you go to its pricing here and maybe it's. I don't, I don't know. Let's. Let's go back into here, right? It's Postgres equal storage. It's charging me a $50 per gigabyte per month. Okay, so I spent 49 cents here. Now it's Postgres SQL access. This is, this is a more expensive thing at 16 cents per hour. Game, maybe so. So my monthly cost so far, by the way, on dozens of apps, right? The total is $18.50. So I'm, I'm like a power user and I'm still only spending $18 here, but maybe Neon database charges 11 cents per hour. And so Replit says, okay, the way that we're going to make money is we're going to make this amazing interface, right, that allows you to use AI and talk and do all these things, and we're going to bring it all together in an IDE, an integrated development environment. Okay? And instead of charging 11 cents per hour, we're going to charge 16 cents per hour. So we're going to make a little bit of margin there. We're going to make 5 cents per hour by reselling the database. Now, for me, that's worth it because I have this crazy amazing thing. And so it's going to charge me 30% more than it would if I, if I put it together on my own. Now I could say, you know what? I don't want to pay 16 cents per hour. I want to set. I want to spend 11 cents per hour. Okay, well then I can go over to Neon and I can sign up and by the way I'm going to pay $19 a month minimum, or maybe zero per month minimum until I get over a certain, you know, whatever. And I can use an API and I can connect this and I can use my solutions and I can find the API documentation and all the things I need to find to make this work or I can just let Replit take care of all that for me. Right. And I'm going to use their manual things. All right. I am actually going to go get a mapbox token so that I can get this going here.

Is that per hour that your site is running, John?

Not necessarily. Yeah. It's more the usage. Right. And so if my website's sitting there, you know, not doing anything, well, I'm going to have my deployment. So my autoscale deployment might cost me 3 cents per day. Right. A deployment means it's just live. But this is a Postgres SQL computer. Right. And so my gen IPI is doing, you know, it spent, you know, 107 compute, whatever. And so that's the majority of it. But if I have this other thing that's live but I'm not using it well, maybe it cost me, you know, 0.33 hours of compute. So it's cost me a grand total of like $0.03 this month to have that live. Does that make sense?

Yes.

Yeah. So I know this is a little bit confusing for some of you and there's a lot of different things in here. The bottom line is these are all the services that are required to run a sophisticated app. You got to have a database, got to have storage for your objects, right. For your videos and your pictures and whatever. Right. Now that type of stuff is super cheap. $0.03 per gigabyte per month. Right. How much have I spent? It hasn't even registered. Right. Okay. Because I just don't need that much. Much. Okay. My auto scale deployment, yeah, I've spent 20 bucks. But that's because I, you know, again, my gen IPI is the vast majority of this 6 million because I've used it a lot and you guys have spent $1,000 a piece to be on this. So I don't care. Right. It's cost me, you know, and, and by the way, I have other services too. And I'll show you the whole thing. I spent $71 on my agent this month. I spent 136 in total this month to make this month alone in April, probably somewhere around $70,000. Who here is willing to spend $136? If you're making 70,000.

Yeah.

All hands need to be up, right? Okay. So don't worry about this type of cost. Your cost early on is going to be pennies or dollars, low single dollars. Except for your agent. Your agent cost might cost you to build this thing. When you work with it for a bunch of hours. Might cost you 60 bucks, 70 bucks, $110. Okay. Right, let's go back. How much did I spend on seek? How much did I raise? Raise? 13 million. How much did I make? About 15 million. Over the course of, you know, that whole time, that's $28 million that I spent to build that thing. Really? I raised a ton. You know, now, now that's not completely true. We paid people, we paid, we paid developers. I, I made some money of that in there. Right? But, but the reality is that's how much it costs to run a business. That's a lot of money if you're spending $100 to build out a fully functioning product that's ready to go, that has a full admin and live courses and, you know, schedules and the ability to go in and, and change, you know, test content and all of my courses and all of these different things that are happening here. Holy cow. Like, that's amazing. So it's, it's, it's this scale thing, right? Okay, hold on. Let's get my mapbox. I want to get my Mapbox token going here. Oh, sorry. I did want to go back to this. Let's say I don't know what this is. It says we need a valid mapbox token. Okay, well, where do I get a mapbox token? Oh, yeah, if you copy this and paste it somewhere, it might copy the markdown. If you paste it into Google Docs. Google Docs. Recognize. What is this? This? I don't know. I'm not a developer. I don't, I don't know what's going on here. Okay, let's find out. So it's going to review my message and it's going to say, okay, great, why don't you go to this website, sign up, and I'm actually just going to click on it and do what it says. I, I know what it's doing, but it's going to be less brain power. Right. Go to the Mapbox website. Sign up for a free account. Once logged in, navigate to your account dashboard. Look for access tokens or API keys section. Okay. You can use it for free. Up to 50,000 map views per month. So I'll click on MapBox as it's marked down and built a document for you.

Right.

So, so anyway you can, you can do that now if I want to close out of this and get back to my chat, I just push X. I don't lose it, right. I can open my sidebar again here. It doesn't disappears my projects. If I want to get back into it, right. I can, I can ask another question and then I can come back to here and open this back up at any time. So I now have this document. Okay. I'm going to say get started for free. And I'm going to. So when you're in the document though, you can push copy. I can just push copy on that thing. So we're going to do that. I want you to say this is a business. Full name Cheney Flow State leadership. Here's this, here's my John at Flow State leadership. John Flow State. Let's do that. My password will be copy your business plan. Okay.

You have to select it first or does it know to copy?

So when you're in the document though, you can push copy. I can just push copy on that thing. So we're going to do that. I want you to copy the whole thing. No, it will copy the whole. It'll copy. If you're, if you have this thing open and you push this little copy button right here will copy the entire business plan. Okay, then we're going to copy your business plan. Okay.

You have to select it first or does it know to copy this?

I agree. Continue. I think I, I think I typed that right. We'll save it. Okay, create my account. Now a lot of time jump into gro, okay. And we're going to start a new chat. Okay. So open Grok, back up, get times when you're setting up an API based account where you're going to need that to your chat and you're going to type this, you're going to say, I want you to do some research on this business plan and provide some feedback. How can we improve it? Okay, that's your problem. It will have you put a credit card in even if it's free. The reason for that is because once you pass the free tier, you don't want your app to stop working. Right? So even if you are on a free tier, put your credit card in. It's not going to charge you. No, it might, it might ask to charge you. You can say no. Just stay free for now. Now with open AI, super easy. And then you're going to hold shift. Some of you may know this already. If you hold Shift and you push return or Enter or whatever, right? Then it won't actually send it. If you just push Enter, then it's gonna, you know, to get to a new line, it's gonna just ask Grok. I want to do some research. And it doesn't have anything. Right? So Shift, Enter, Enter and you're gonna paste your business. If you're doing an AI based feature, you will have to put $5 of credit on your account. And that's fine, right? You're building again your million dollar plan, okay.

As well.

Yep, yep. You've spent a thousand dollars to learn how to do it. It's going to cost you $100 to build it. You can spend the shift. The shift return works on Mac as well. So you're going to write this question. I want you to do some research on this business plan. Present VTech$5 to get your feature working. Because you have to, you have to do that. Okay? So before you do that deep search, okay, click this Deep search button. Groq has some neat tools called Deep Search and Think and it even has something called deep Deeper. So I'm now going to have a verification email. We've already done this. We're going to activate my account, verify my email search, which we won't try today because it takes longer, but it's now going to check the Internet for a hundred different articles on whatever. We're going to figure out how we can really stand out here. And my billing information. Right. Wants me to do this. Right. So I'm going to go ahead and click Go.

Would you leave?

Sal 6200 West. This is my, this my tax ID. I can do that. Optional. I don't need to do that right now.

Within your open questions or would you take them out?

Whatever you want. Whatever you want, right? Whatever's in the business plan, just paste it. It's going to do a lot of thinking. It's going to do a bunch of research. I'm not going to update that right now. Okay. So I didn't even put it. We're not done with that business plan, right? And so we're going to, you know, we'll dive deeper in. But look, it's already searched 39 sources. It's been, it's 39 sources. It's been 20 minute card in. But now we need a token. Okay. Here's my default public token. Okay. Use protected tokens from. I have applications. Okay. So let's just, let's just create a token. Okay? One seconds. And so now it's researching weather apps, then it's enhancing the business plan, evaluating an AI integration, exploring AI benefits, is trying to figure out exactly how we're going to make this, you know, even better then. Now, if I'm building a business, guys, I'm going to read all of this. Yeah, Every word. Now, there's all kinds of stuff here, right? Styles, tiles, read fonts, read data sets, read secret scopes, right? So here's what I'm going to do. I don't. I don't know what I want here for my token, okay? For the purposes of this course, we don't have time to sit here for 12 hours or my name is going to be Nimbus, right? And that name my app. Or, you know, not that it would take you 12 hours to read this, but it will take an extra. It took me, you know, 34 hours over three days to build Gen Aipi. My. My business that I built in three days. But that's because I went hardcore for. Right? Didn't I call it? Isn't it. Isn't it Nimbus? That's where it is. Yeah. I'm gonna. I'm gonna verify because my brain is just not working very well and I'm gonna come back to here. Weather app. Weather app here. I've got my three days, right? And I read tons of things and I really put thought into it, and that's why I have now generated a huge amount of money with it, is because my business, Nimbus Development, I'm gonna. I'm gonna verify, okay? That's what it's called. I generated real ideas that were going to turn into money. And I really put my thought into it and went back and forth and. And not only that, I wanted to understand it, right? I care about what I'm working on, and so I want to understand. I'm going to call it Nimbus Now. I don't know what I need here, so here's what I'm going to do. I'm going to go like this. I'm just going to screenshot this and it. So you will want to do that at some point. For now, you're going to go ask it.

Boom.

And you're going to go back, you're going to read later, okay? Read your business plan later. I want you to understand what's in your. What should I select here? Business plan. But we don't need to, you know, be perfect experts on it right now for the purpose of the course.

Yeah, sorry, old guy here. So in.

Looking at this, it's going to say, okay, you know what.

Is there a way for me to save my post other than just clicking on the little save link and save the results of this, the search so I can find them?

Actions that are already selected is all you need. We're great. Or it's going to say, hey, why.

Don'T you select later when I want to read all this.

Yeah, so your, all of your chats are saved in Groq, right? So you have, there's a history button and you can look at chat history, you can go back like this, this, this and this. And we'll have exactly what we need. Okay. So again, we're kind of getting ahead of ourselves with this API thing that we did today. We usually do this tomorrow, but you guys will be able to look at any previous chat you've ever had, just like in chat GPT. So you'll be able to go back and find all of those. All of those things. Okay, so there's a bunch of different things here we're working on. You know how to do this financial aspect, operational risk management, leveraging emerging technologies, marketing, community building. There's a lot of things, a lot of iggy that much further ahead. All right, so it's doing a little bit of thinking. Come on. My, my life is buffering. All right, so here the mapbox token creation examples and things that it's figured out. Okay. Overall process, all right. For public purposes is great for under secret scopes map read. This allows you to access the map. Okay. All it says it's a strong plan, but could improve a little bit in market analysis, financial projections and AI integration. Right. This is where we think we need to do some better things here. So now we've got some feedback. Okay, Grot gave us some feedback on our chat. So we're gonna do map to read right here. So we're gonna let it do that and GPT plan by doing a bunch of research. Right. And here's all my citations. So I'm going to come down here to the bottom here at Grog. I'm going to push the copy button. I'm going to copy what it just create token. Okay. You use restrict did. And then I'm going to bounce back in here into chat GPT. I don't need to do that. And I'm going to type or I'm going to paste right here. I'm going to say. Well, first I'm going to say URLs format is important. Yeah, I don't think I need that. Okay. Password. Please consider this research below and refine the business plan. Okay. And I'm gonna do my shift enter enter thing. I'm gonna. It's gonna ask for your password. Might ask for a 2fa. It might ask you to go verify your email or something. Yeah, I might have literally typed my password incorrect paste back from Grok into here. Right. All this information go. Okay. So now it's taking everything that submit this. Yeah. Literally when I set it. See that, That's a bad. That's a bad UI there. Because it didn't ask me to verify. Just took, you know, a minute to do and did a month's worth of research and it's going to bring it back into here and it's going to say, okay, let's make this even. So I don't actually know what my password is now. What I'm going to do is I'm going to stronger.

Okay, so would you suggest then that if. If Grok is still spinning, has not produced anything, that something probably went off the rails?

No. Well, what's up? Share my screen really quick and I'm going to look at my password just to see what it says it is.

It's another vote for password managers.

Yeah, totally. And you know what? It's crazy.

Took a couple minutes for it to actually kick into gear. I just waited and it kept it.

It finally worked. Yeah. Guys, Grog.

Second from the second you said it.

I. I see. It's exactly what I typed. Somehow that didn't work there. So I'm going to return to the sign in. I'm going to go back here, I'm going to try it again.

Did it and it's still not.

Is it just those three little dots that are spinning? Yeah. Sign in. I actually did type it right, which is crazy. So I don't know what this is a UI issue, guys. Grok is having some problems and so what you're going to need to do is just refresh the page. Just try refreshing the page and see what happens. It will not lose the results of what it did. It will probably just start displaying the results. Maybe I. Maybe I typed it wrong here. I did this. So let's go back to this here.

John, did you copy everything from key points below on?

I just copied the whole one more time. Create token. It's going to ask for the password. No, whatever logged me in. Somehow I had lost my authentication. Somehow that happened. Thing just copied the whole thing. Like it. Yeah, it can parse it. It understands. Okay, so this is the one I Created even though I thought I called it, you know, and said it's a response.

Well, like did you open the bubble that had everything or did you go below?

Oh, yeah, yeah, yeah. So good question. Yeah, so I didn't, I didn't go here and you know, like have it like really show all of the thinking and all of this stuff, right? Like that's a lot of thought, whatever, but doesn't matter. I'm going to copy this token. Okay. Usually that's that button right here. They usually provide a copy because it's a crazy long, long thing here. So you guys. No, I just want the synthesis, right? So yes, it is key points and everything. Know my API. So if you wanted to steal my API credits for mapbox below, right? So I just come down here and I just push copy, right? And that will copy from. From. You could come here and you know, copy this and paste it into yours and you would then be stealing from my AP points right there. And so yeah, if I just push this little copy button, paste it back in. Now I've refined my business plan with some amazing research. Now RPI usage and charging my credit card to use a map based function. I trust all you guys not to do that. That's why I'm showing you this. But in real life you don't want to show people your tokens. Okay. All right.

Yeah. For, for me.

Does that work for you?

No.

And that's a security feature. Not that they're going to, you know, hack you and.

But I pasted it, I pasted it into the same thread where I had been doing the.

And like steal sensitive information or steal money from you. The only thing they're going to steal is.

The JavaScript program before, so it may API usage.

Right? And so if you see your API usually like go off the charts, it means you did something bad and you accidentally did that. So now I can come up here and I can drop this in. That's my mapbox token. I'm gonna click continue. I'm gonna click add to account secrets. So now it's saved so I can use it on future feature. If I ever doing something with Maps in the future, I could use that same token if I want. Okay, now it looks like nothing's happening here because that thinking thing wasn't there. Sometimes the UI disappears, just like happened on Grok, right? Just click refresh. Now the cool thing is if I click refresh on Replit, it's going to keep working. It's not going to mess with anything. It's Going to keep doing it. It's going to refresh the ui. Right? That's great. So now we're going to come back to our preview here. I'm actually going to open up a web view. Where on earth is my web view option? My web view isn't showing up here.

Isn't that Preview?

Well, no, usually Preview is doing, is showing like progress. But maybe it's Preview on this one. Maybe they changed it from WebView. I've, I've had that. Or you know what, maybe it's because we're building an app and it's the type of system that it's building. So that's okay. We'll let it do this. Look at Preview. We'll refresh it. Yeah. In this case, Preview is acting as webview, but sometimes it's webview.

I do see a webview on mine.

Yeah, exactly. So why that's not showing up here might be because of the type of project that I'm working on right now. I'm going to click pause. Actually I'm going to ask it. I'm going to say, are you building an app that is web, but that is a web app, web compatible? Just going to ask it right now. So it's going to go through some guys. The first time I built this Echoes app, it took me 30 minutes to get past errors because it's trying to figure out how to do something really complicated. And yes, it says NIM is a fully responsive web application. That's great. So I don't know why. I just wanted to verify to make sure that that's why or not why I'm not seeing this. But yeah, maybe they just changed my UI to Preview instead of webview. I don't know why, can't answer that question. But we'll sit here and we'll let it keep doing its thing here. Now it's freaking out here right now and I don't see anything. So what I'm going to do is I'm going to click refresh here and I'm going to say resume. Keep building. I'm also going to show this still. Okay, so guys, this, this as we get to this part, by the way, we're done kind of with our, with our hosting and deployment. We kind of just kind of, we're done with the, the coding. Quick start guide. Really just front end, back end, UI databases, APIs, authentication. Just a few terms that will help you be a little bit more conversant as you're, as you're working here. Right. So it's still trying to figure out this map visualization. Still trying to figure this out. We've got the mapbox token. We're still trying to figure this out. So I'm just going to keep going until I'm getting to something where I can actually edit. Okay. So if you guys aren't already doing this type of stuff already, you should be, you should be starting to talk to it. You might not have errors, you might already be into your system and saying, okay, how do we make this work better on day one, let's go back to this. And maybe we didn't even, we didn't even do this in the beginning, but on day one, this is what we're going over. We're going over the business idea, right? I focused on an AI powered business idea. It's an it's AI enhanced ChatGPT Grot to come up with some ideas. We're going to define the minimum viable product. Okay. This is really how to build a business in three days. That's what it's trying to do. But as we keep going here, step two, you're going to set up the business infrastructure. You're going to register a domain. If you're actually building a business business, you would do that. You'd set up your email. I would set up Google Workspace, right? I like to use Google Workspace or actually it's called something else now. I think maybe, maybe it is workspace. Yeah, it is workspace now. It used to be called. What did it used to be called? I don't remember. Anyway, but you could set up an email address, right? So I could set up something@nimbusweather.com Right, if I wanted to do that. So for my Echoes app, I built out a full business for that, right? I have a full business for gen aipi, right? So here's my admin, right? I came in here and I built out john aipi.org I built that and I bought the domain and I connected it all and I'll show some of that stuff later. We'll work on that and we'll answer some questions on those things tomorrow. We don't necessarily need to do that today. Unless you want to really build it out. Unless you're like, I am. You don't do this guys unless you are completely committed to building this business. That's when you do this. When I built Genaipi, I liked my idea and I was like, I am building something that's going to help people get better at AI. I'm going to build this Gen AI PI, I love it. I love the business idea so much, I said, I'm gonna do it right now. When I first built the business, I did not have this live courses thing that you guys are on right now. This was not part of my business plan. It wasn't. You go back and you read my entire ChatGPT history related to Gen IPI, right? It was. The Build a business course was later. Right? Building a business with AI. This was later. Okay. My initial idea was I want this AI aptitude test platform. Okay, here's my very, very first prompt. When I went to build my business, I said, I want to create a website that checks the proficiency levels based on an intelligence test that shows the aptitude for an individual to succeed in a world with AI. This means they need to know how to talk to AI, how to be good prompt engineers, and they need to have a good brain to figure out how, how complex systems can be to put together. Can you help me build this? Okay, that was my first prompt On Thursday morning, February 28th, when I went to build this. Okay, so, okay, great. Let's build an intelligence test. So we're going to need multiple choice, scenario based interactive challenges. Great. We're going to need proficiency levels, interactive AI challenges, a dashboard and insights, gamifications. That was kind of my first idea. I said, let's start with a good name, right? How about AI Aptitude Institute? AI. AI Thoughts, Better ideas. And then I got the aipi. I like that better. AIPI was better. It flowed better. And then I realized that wasn't available, so I ended up having to land on gen aipi.org, right? So I went through that and went through this whole process and figuring that out, but I started with the name, started going deeper, but that was my first prompt, right? And I just built this out and at that time I realized I really wanted to do this. And so I figured out my name. I focused on, you know, someone asked me earlier, you know, do I need to figure out the name right now? Well, if you want to. Really? If you want to get the website, yeah, you need a name, right? And it took me a while to figure it out, but after a couple hours of talking back and forth, right, this is my full chat, right? I, I figured out, okay, And I read all of this, right? I was thinking through it and I was checking domains and I was checking things and I was like, okay, let's roll with this. Okay, so if you're actually building a business, yeah, it does matter. See, where are we? We're on flow, state leadership and we're on this right here. So today, you know, day one, yeah, you're setting up the email, you're registering the domain, you work on some logo generation. We're going to do branding and logo stuff tomorrow as along with some of our more visual stuff, you know, image generation, maybe even some video generation. Although we really like to do some of that maybe on. On the third day, social media, online presence. Right to you got to build out an Instagram, a LinkedIn, Facebook X. Really what you're trying to do, guys, when you're building a business quickly or building a business at all and working on this, on this, you know, this app. Yeah, let's see if this is actually working now. What you're trying to do is you're trying to get people to not have a reason to doubt you. Right? Yeah. We've become unresponsive here. I can wait or I can refresh this. I'm going to try to refresh it, see what happens here. See if my app is actually loading now. Hey, Nimbus, that's cool. What is this? I just see Nimbus and I'm super intrigued now. Is the map loading correctly? No, it's not. Do I see weather layers? No, but I see this. It's like pulsing. Why? Is that an accident or is that on purpose? I'm just.

John.

Yeah.

Question. Once you're working in replit, when does it make sense to go back to Grok or. Or Chat GPT? If you want like let's say you want to change your name from Nimbus. Do you do that in replit or how do you go back and forth from the other tools?

Yeah, it's a good question. So one of the things that we haven't done yet. Oh, thank you for saying that. That actually reminds me of a super critical thing that I. That I really care about. So this is. Yeah. And welcome. It looks like Michael left for a minute and he's back now. So yeah, I've got switch to my phone for my pickleball league. Just gotta drive there. So. Okay, great. I'm glad you're here for this because it's important. Right? So it's a great question. The question was asked, you know, why or you know, now that we're in replit, like do I keep using ChatGPT? And the answer is absolutely yes. Right. I have done so many things in here, you know, I'm still. Remember this is my project manager, right. Chat GPT is my project manager. It's Going to keep me on task. Now we also. I'm going to go back to my weather app here and I've got my business plan and I've got my development plan. Now this is a critical thing inside of Replit. So first of all, I got to figure out what's going on. I'll get back to that at some point. But what I need is. Yeah, it's the login with replit button visible in navigation. Well, I don't want to actually log in with replit. I just want to create a login with the database. So we'll worry about that in a minute. But we need to create two documents. Okay, we've created two documents, but we need to bring them in here. We've got this development plan. Okay, so I'm going to click on this. I'm going to copy it. Sometimes it fails to copy the first time and I just have to do it a second time. Time. But I'm going to paste this in here. This is my development plan. Please create a progress MD file with the attached development plan. Also create a readme MD file with the attached business plan. Okay, I'm going to copy the business plan for this other thing here. Business plan and copy. It's going to fail and I'm going to copy again and it's going to work for whatever reason. Just learned that little quirk when I'm using this is when I'm using the. The app, when I'm on the website, it doesn't do this. But when I'm using the chat GPT app on my computer on Mac, whatever reason, I click copy the first time, when I open up a document the first time, it fails. And then when I click again, it works Anyway, little thing there, you've got this readme MD file and this progress MD file. This is one of my tips and tricks. This is big time. It's very important. Okay, I'm going to paste these two things and I'm going to ask it to create this and I'm going to click go. Okay, I'm going to. I'm going to not worry about this right now. About what it's. What it's doing. I'll get back to developing in a second. But right now I need it to create a progress MD file and a readme file. What this is going to do is it's going to again, it's going to give very specific instructions. Now we already, we already provided this content. It's actually in the attached assets. Right? I actually pasted it again. I didn't need to. I could have said look at the attached assets and use the business plan and development plan to create these. But for now, In fact, I probably could have just said this. This thing here. I'm going to give you guys this prompt here. Okay? Let's put it in the meeting chat. Please create a progress MD file with the attached development plan. Also create a readme on Me file with the attached business plan. Okay? Those documents are already in your attached assets. If you click on your project right now and you open your files, you can open your files right here in this top left thing, okay? And you open attached Assets folder, you'll see your pasted thing. So just paste this thing right here and it will create it. So it's going to create this progress in me file and then the second will create the readme MD file. And you're going to keep referring to these back and forth as you build this out. Okay? And I'll show you how to do that. So while this is building, to fully answer the question that was asked about how do we use ChatGPT now? How do we change our name? Well, yeah, you're going to Update the ReadMe MD. You're going to say, hey, please update the ReadMe MD and instead of Nimbus, we're going to use Stratus, right? And it will update that entire document. And you'll go over to Chat GPT and you'll say, hey, instead of Nimbus, right, we're going to use Stratus. Please update the business plan document and the development plan document. And it will be like, okay, and it'll update. It'll update. Done. Okay. Or hey, we need to update something here. The other thing you can do is you don't have to necessarily, like, you don't have to update every single document. Every time you make a change, you can say, hey, we're creating a new document, right? I am going to update the name from Nimbus to Stratus. And I could go and I could just say, boom. And it will say, okay, great. It will update its memory and it will know that inside of your project. So after you have, you know, if I were to come in here into my Gen API and say, I'm changing the name from Gen AIPI to awesome AI, I can just tell it that it's going to update its memory. The project will know, okay? The most recent name is awesome AI, not Gen ipi. And it'll just know, okay? So just tell. You don't have to go through and necessarily update everything now in repl it. Yeah, you're going to be like, hey, you know, we need to, we need to update this. And so anywhere where you see the name, anywhere where the name is set, you know, it says Nimbus, we need to update it to Stratus. We're going to do that. So right now, though, so I've created. By the way, let me show you this. If I click on readme md. Look at this. Nimbus, the world's most beautiful and intelligent weather map. Awesome. This is the whole thing. Competitive landscape, everything that we built. Here's my progress, here's everything that's been built, everything it needs to do. Now, here's a cool feature. Here's what I like to do. I'm going to actually ask it one more thing before I, you know, discover what is happening here, because I'm still super intrigued. What is happening? Like, what has it built? And why can't I see anything and let me into this awesome thing like, like it has built something that I'm really excited about. All right, here we go. We're going to update this progress thing to be a little bit better. For the progress MD file, I want you to create a. And I'll copy this prompt so you can use it. I want you to create a checkbox system that shows what has been completed and what it. What has yet to be completed. Please update the progress MD file to reflect the current state of the app. Copy. So, and I share this with you chat pastes. Feel free to do that if you want. But then it's pretty cool, right? And we can actually watch it live. I think it will update it live. I'm not sure if it will. I don't remember. It's going to update it. It's going to review it and then it's going to. Actually, this is going to take a minute because it's going to actually now look at the entire project is going to assess everything that's being created and it's going to update it. This might take a minute.

John, this is like a silly question.

Go ahead.

We've been listening to you for a while. Like, should we be on the side messing around with what it spit out at us or.

Yeah, that's up to you and I. I addressed this earlier before you knew that it mattered. But yes and no. If you're really good at listening and doing. Yeah, great. Okay. If you just like watching and learning and then going back and doing later, then do that. Right. And so this is true for Basically the rest of the course, I, at this, by this point, like, I know I work fast, guys. I move fast and I teach fast and I work fast. And, you know, you can decide whether or not you want to try to keep up with me or whether or not you want to just kind of watch and learn and kind of work a little bit on the side, but really, you're going to dive into this after the course. A lot of people will, will keep up with me. And that's great. Amazing, right? Other people will say, okay, I'm going to watch this and then tonight I'm going to work on it. And you're going to work on it till like 3am and you're just going to keep going because you just, your brain won't turn off. You want to keep going. And then you'll wake up tomorrow and be like, oh, my gosh, I'm so tired. Right. So yeah, that's a preference thing. How many of you guys have been kind of going along with me and kind of trying to work on it a little bit? Yeah, see, John's been going, Kurt's been going. That's fine. Great.

And I got, I got app that you can sign up for right now.

Awesome. What do you got, John? Do you want to share? Do you want to show something? Got something Cool. Yeah, share your screen. Let's. Let's see what you've done. I'm super cool.

I don't know about everyone else.

No. Yeah. Again, this is just one of those cool things that is different for everybody. Hey, neighbor. Exchange goods and services. Amazing. You can create a new account that's super cool. Now we're going to check something in a second to see if it's actually working well. But go ahead and create the password, whatever you're doing here. Yeah, I love seeing, I love seeing what you guys are doing here. Okay. Agree. All right. Registration failed. That's fine. Okay. Something didn't work. Email already in use. Yeah. You've already created. Yeah, someone has already created it, so you might have to do that. So just go ahead and log in. Yeah, go ahead and. Yeah, just use your save. Yeah, you can just click it's already saved. Yeah, log in. Done. Hey, welcome back to hey neighbor. Cool. So profile. Yeah, so yeah, you've already created something that works. This is awesome. I love it. Yeah, I want to check something. John, let's check some. Some critical things. This is a great learning opportunity and very important for day one. When you are building, I want to look at your database. I want to see if it exists.

Okay.

How do you.

What do I need to do that?

So we're gonna. Yeah, we're gonna do a new tab there. We're gonna click database. So we're gonna. Yeah, we're gonna do a new tab there. We're gonna click database. Awesome. Now this worked. Click see my data. Go ahead and click see my data because.