Cohort 10 - Day 1 - 5-12-2025

And, and I hope nobody else joins because we've got a nice 3x3 here, and that's a nice grid. So welcome, guys. Welcome, Travis. Um, so we'll go ahead and dive in here. For those of you that don't know me, I'm Jon Cheney, and I live in, uh, West Mountain, Utah, which is on the south side of Utah Lake, for those of you who don't know. And I'll do a little bit more intro about myself soon here, but really excited to have all of you guys in. And, uh, we're gonna be doing a lot of really fun stuff. We're gonna be moving very fast. Um, note that this is recorded, so if you're like, oh my gosh, I didn't catch that, you know, don't feel bad. Like, speak up and say, hey Jon, I didn't catch that, can you say that again? I want you to say that during the class, but know that you can absolutely go back later and watch the recording of it. I will, I will send the recording out at the end of each day, usually by, I don't know, a couple hours after the course ends, because this is a 4-hour thing. It takes a long time to process and then it has to render and then it has to upload. And anyway, that's the process on YouTube as well. But I send you a link to all 3 and then we'll send them at the end again. So don't worry about missing something. This is supposed to be very, very interactive. Every single one of you will share your screen. That several times during the course. I'll just pick on you randomly. I'll be like, Jorge, you're up, right? Travis, you're up. Let's see what you got. Kelly, you're up. So, so be ready for that. And by the way, we have two Kellys, so that's going to be a little, a little rough, no pun intended. Kelly Trust is actually my sister, which is cool. I've been doing this as the 10th cohort and she finally said, hey, I need to jump in on your course. I said, all right, go ahead and join. And we'll call you Wally Kelly or something like that. I don't know. I'll do my best. Kelly with an I. We'll figure it out. But excited to have both of you on here. I think this is the first course with two women in it, so that's cool. That's a, that's a nice thing, right? There's so many, so many dudes take this course and that's awesome. We like the guys, but, but I love it when we get the women on. They add a different aspect to it and that'll be fun. All right guys, let's start really quick with just a basic intro from each of you. You get about 42 seconds as your time limit. I just want to hear your name, where you live, what you do for work, and then I want you to just answer this one question for right now. Have you ever written a line of code before? That's all I want to know for now. We're just going to go in the order that I see it on my screen. Which I think is the order that you guys joined. We'll start with David. You're on mute, sorry. Make sure you guys unmute when you do this. We'll get good at this after 3 days.

My name is— sorry, I live in Mexico City. My name is David. I have very basic experience coding. I took some Java, just one Java course. At college, and then I realized that I was not going to go into information systems.

Awesome.

Um, what was the other thing you wanted to know?

What do you do for work now? What's your— what's your—

I was, I was working, I was working for an energy consulting company. I left them at the end of January, and I'm trying to figure something out. So I thought this could be, you know, what I need to, to get off the ground.

So I love it. Amazing. Thank you, David. Um, Mike. Andrea, Mike.

Yeah, my name is Mike. I'm a home health physical therapist. I have absolutely zero coding experience. It was difficult for me to get on this Zoom call, so we'll see how this goes. But I'm excited.

I just—

even messing around with AI was cool. So I'm sure everyone's ahead of me, which Sounds fun for me too, to be in the class. So I'm excited to take the class and be here with people, hopefully smarter and clearly more advanced than I am.

Now, you're doing awesome. What, what do you do for work?

Like, I'm a home health physical therapist.

That's right.

I see they're all homebound people, a lot of joint replacements.

You go to their homes to do stuff.

I go to their homes. So people always ask me sports injuries and I'm like, I don't know.

Well, maybe you can come to my home.

Yeah, yeah, I saw that. Biceps tenodesis is my guess.

Good, good job. Yeah, um, my, uh, yeah, my, my biceps tendon right here, um, had a high-grade tear. Kelly, they tore down low. Yeah, and, and, uh, anyway, it was actually way worse. It was actually way worse when they opened it up. Anyway, so I got that just, I don't know, 9 days ago. I get my cast off, I get this off tomorrow. So I'll have— it's tomorrow after the class, so my last day you'll be able to actually see my forearm, which might— you might not want to see because it's probably terrible right now. I haven't looked at it in 10 days. So anyway, um, awesome. Thank you so much, Mike. Kelly, go ahead.

Oh my goodness, I put some nut butter in my mouth right when you said that. Um, so I, um, and I'm an athletic trainer for bodybuilders, so I do a lot of body work for them, a lot of physical therapy with them, but minor stuff. Cupping, Graston, just got certified in dry needling. So after, uh, I went to BYU, graduated in sports medicine there, so my own business is Sports Medicine Kelly LLC, but it's still new. It's about 5 years in the business right now, but I primarily work with bodybuilders, but I do kind of all-encompassing health for them, and I've been getting into managing said athletes. So needing to figure out how to write contracts and, um, actually have things locked in to be profitable for me, a little more profitable. Um, as far as coding goes, in my 11th grade, uh, DICOM class, I wrote code because we had to build a website, but that was like colors and lines and 3 different options for a website, you know, so very basic stuff, and I forgot most of it. I know there was a lot of slashes and numbers, and that was the most that I got out of it. But I know that there are things that write that a lot faster than me now. So that's kind of why I'm here.

That's awesome. And you do some bodybuilding yourself too?

I do.

I know you just— I'll brag for her. She just won a first place in a competition, I don't know, a couple weeks ago. So great job to Kelly. Um, anyway, yeah, good, good stuff. Uh, yeah, I love it. That's, that's really fun. Drew, go ahead.

Yeah, Drew Ludlow, Live and Holiday. Bounced between startups and politics for the last 15 years and just been working on a startup that actually just sold last week. And so figuring out what to do next. I've worked with a lot of technical people. I'm usually the one, so I know a lot of engineers. I'm usually not the one who is technical. I know just enough to get myself into trouble. I've gotten done a few things in Terminal, which is like the place to code on Mac and get myself into trouble when I try to do stuff. But I've made some attempts and I've played a little bit actually having ChatGPT coach on that with mixed results. I'm excited to dive in more.

Very cool. That's awesome. Love it. Kelly, Kelly with an I.

That one, Ellie with an I. I live in the beautiful Kamas, Utah, up in the mountains, and I currently work for Wally Cases in the operations department. Connor and I are coworkers. I wanted to take this class and I asked for the time off, and then my boss said, well, why doesn't Connor do it with you and we'll pay for it? So here we are. Yeah, but I have big dreams, so this is just one step in helping me with my big dreams, and I have never coded. Anything in my entire life. I know absolutely nothing about coding.

I love it. Kelly, how did you find out about the class itself?

Um, from you. I met Jon 20 years ago but have not talked to him since, so it's great to talk to you.

Yeah, for sure. Awesome.

Facebook.

Yeah. Yeah, no, definitely. I love it. I love it, Kelly. Uh, Connor, go ahead.

Yeah, so with Kelly, I also work at Wally Cases in the operations. I live in—

Is he breaking up for anybody?

And I actually, I have a degree from the university.

Hey, Connor, you're breaking up a little bit there. If you can hear us, we're gonna go like this.

Come back to me. I'm gonna move.

Okay, perfect. Yeah, something's wrong with your signal. We'll come back to you. Um, thanks, Connor. Um, I'm sure he— I was about to find out where he lives and I was so excited. All right, um, Travis, go ahead. Yeah, I'm at a—

I just live in Draper, Utah. I'm at a startup called Boostly. We're here in Lehigh. Um, I'm just, uh, helping do whatever I can to help the company grow. Um, and I have very, very little experience coding. I've written a few lines of HTML like 10 years ago when I was writing some email templates, and that's about it.

Awesome. I love it. Okay, cool. Thank you, Jorge.

Yes, so I am originally from Utah, uh, living in Florida. Apologies for the lighting. I'm traveling, so the lighting is terrible. Um, and, uh, I've done sales for a really long time, primarily in tech sales. And I haven't written really anything in code. I think the only time I wrote code was when, um, what was it? Not Facebook. Um, what's the other, uh, prior to that? It was like all the different— you had to like your friends and stuff like that. MySpace. That was the only line of code I wrote was like for customization on MySpace.

That's it. I want to look up and see if my MySpace profile is still there. Sure, I'm sure it is. I haven't taken it down, but anyway, that's awesome. Thank you, Jorge. Yeah, very cool. And then Connor, you were about to tell us— you were kind of halfway through— you're about to tell us where you live. You said, I live in— and then kind of broke up.

Yeah, so I work with Kelly at Wally Cases, um, in operations. I live in Ogden, Utah, and I actually have a degree in computer science from Utah Valley University, so I have quite a bit of experience coding.

See, that's part of what I want to learn. It's part of what I love to learn is who is actually a developer in here. That's helpful because we're going to go over some things. Thank you, everybody. I am not a developer. I couldn't write a line of code if my life depended on it. But with AI, I've written probably close to 3 million lines of code now across 20-30 different applications. And they work. Things work. I click on buttons and things happen. As a user, it appears it's working. Now, I know there's a lot more to it than that, Connor. There's all kinds of things. We'll have a little bit of discussion about this, but some developers say, "I can't code. It makes mistakes." I'm sure it does. So do humans, whatever. But the reality is some of the things we're going to be learning today and over the next couple of days are going to be all about rapid iterating, rapid prototyping, rapid MVPs, getting to your first dollar as fast as humanly possible. And I can promise you that you do not need to have a perfect product to get to your first dollar. You just don't. It's a little bit different with Wally Cases, for example. You guys create, if I'm, if I'm correct, they're, they're phone cases that are like a wallet as well, right? I think my wife has one. Yeah. Awesome. Awesome product, right? It's tough to build a physical product because you have to actually get it kind of right before you can get it in front of somebody, you know, and it has to kind of be in at least close to an end state before you can put it in front of somebody. But software, as long as it does what it's supposed to do, you can start to get it in front of people really, really quickly. And lots of times software is just a website, you know, for promoting a service that you do, you know, Kelly or Mike, you know, or different things like that, right? Maybe you need to build some booking software. I need to build this and that, whatever things that, you know, payment, payment things. Now you need to make sure it's secure. You need to make sure it works. But ultimately your goal is I want to get up fast. I want to start pushing it out there. I want people to be able to sign up and use it and, you know, build, build a real business. So, so we're going to be doing a lot of really, really fun stuff. And you guys are all going to learn how to code with AI at a very, very high level. My estimation is that when you're done with this course, you'll be close to the top 1% in AI users globally. Right. A lot of people, most people out there that are using it, there's a lot of people that have ChatGPT downloaded on their phone, but they're using it as like a glorified Google. Right. And, you know, they look it up and that's cool. And, you know, maybe they'll ask a follow-up question. Maybe they'll generate a Studio Ghibli picture or something fun. And, and then that's about as far as it goes. Right. But AI can be used to accomplish just about anything. And if you know how to use it and partner with it, right, that enables you not in like a Tony Robbins rah-rah-rah style, but legitimately enables you to do just about anything, right? And so, we're going as technology continues to improve, right? I mean, if you don't know how to do something, it used to be like, "Oh, there's an app for that. There's a website for that. There's an app for that." Now, there's a YouTube video for that. If you need to learn how to change your toilet, you've never done it, you can watch a YouTube video, and now AI can walk you through step by step. And as it continues to improve its interface, it will get even easier, right? You'll be able to just use your voice more and more to talk to it. Maybe eventually, not maybe, but at some point there will be, you know, you'll have a robot in your house if that's your style. If it's not, if you don't trust that, then that's fine. I will order the robot when it becomes available. That's just, that is my style. I'm an early adopter. Tech guy. So, a little bit about me. I founded a bunch of companies but the big one that I found was back in 2016, raised $13 million for it. It's called Seek and ended up selling that, exiting. I left the company at the beginning of 2023 in January and put a CEO in place to keep running part of the division. We kind of did a two-part sale. We sold half of our technology to one group and the rest of it to another group. But that is completely, completely exited now at this point and it is all done. And in that process, by the way, guys, raising capital and going and delivering on a big tech company is insanely difficult. Travis is in the middle of a great company that's doing that and it's really hard. I talked to Shane. I've talked to him a lot over the years and it's stressful, right? It's hard. What's your role there, Travis? Are you a co-founder or?

My title is founding team member. I do whatever. I led our sales team, I grew our CS team. Now I'm focused on some AI enablement. We're pretty bullish on it. So whatever the day requires.

I love it. I love it. Well, that's cool. Well, all the things you're gonna learn in this course are gonna make you even more of a generalist with AI because you're gonna be able to do anything, right? Someone will say, hey, can anyone know how to do this? Nope, but I know how. I know how to figure out how to do it. And so that's really the goal here. We'll be going over probably 15 to 20 AI platforms during this course. We only have 8 hours and 40 minutes left of the course, and I promise you it is going to fly. I had somebody take a course from us or from me on Friday. I did a corporate training for a big group of people, and the CEO messaged me after. He said, I have never been so like engrossed for 6 hours straight as I have ever. Like, I can never sit through trainings like that. He's like, I wanted it to keep going when it was done. So, so hopefully we can have a lot of fun. So we're going to start right now. I'm going to do a group poll right here. How many people have signed up paid version of ChatGPT? Raise your hand. Okay, everybody. Awesome. What about Grok? Same question. If you have not signed up for Grok, it's okay.

I signed up for Grok, but it was— is it free? Because I didn't—

Yeah, yeah. So you can— as long as you sign in for an account, it is free, and then you can upgrade to Super Grok. For this course, I'm going to make that one kind of semi-optional. I like it because there's a few features you can get. With, with SuperGrok, more context, more capabilities of deep research and things like that. But we can work around it if you don't have it. And what about Replit? And I'm going to ask this on the opposite. Who has not signed up for Replit? Raise your hand. Okay. 1, 2. All right. Perfect. So just a couple of people. You're going to need Replit. So make sure you get that done. The instructions. I think there's a little code I gave you that gives you $10 of credit when you sign up for it. It's $35 a month. And you can use it for the course and be done with it, or you can keep it forever. I purchased the annual— the annual one is a better deal monthly if you're planning on using it longer term. By the time you're done with the course, you might say, oh my gosh, I have to always have access to that. But I started using it close to a year ago probably now, and it has changed and improved dramatically since then. It used to only be $180 for a year, which was awesome. Anyway, they've changed that a little bit. Guys, we are going to dive right in. We're going to start today with a coding exercise. By the way, this course is not really about coding. It's about AI, but it's also largely about coding because that's what we can do. It's about getting results. That's what we're working on today. It's all about getting results. Here's what we're gonna do. This is really, really easy. I'm on Grok. Why don't you just watch me for a second? And then if you're really good at watching and doing at the same time, great. Otherwise, watch and then I'll give you a second to just try it, okay? And we're gonna have every single person in this exercise share their screen. So I want you all to make sure you know where the share panel is on Zoom and make sure you know how to do that. Okay, so we're gonna start with this. We're gonna use a language called p5.js. And p5.js is what is called a front-end language. Front-end means, you can see this picture and you can see this place where you can type and you can see buttons and colors. That's frontend stuff, things you can see and interact with. That's all you need to know. This means that it doesn't use a backend or a database. We're going to do everything, we're going to render all the logic, it's going to be on the frontend. Here's what we're going to do. Let's just start with create a mind— let's see if I can type with my cast on— mind-blowing visualization with p5.js. All right, let's just see what happens here. All right, I didn't give it any instructions really. I basically just said, hey, you know, make this for me using p5.js. Okay, so there's a little— on most AIs, there's a little copy button that will float over code, okay, and then some other places. But right here, I can just click copy, and I've now copied all of this code to here and then I'm gonna click on this p5.js web editor. I'm going to delete this and I'm gonna paste it and I'm gonna click play. Okay, all right, boom, I've coded something. That's all it took, right? And I didn't give it any instructions and so it's kind of random. The AI is just gonna say, okay, I'm gonna try to figure out something he likes, right? So I'll do this. It looks like, let's see here. Yeah, it doesn't look like it's really interactive. I'm trying to see if it's wanting to follow my mouse. It's not really doing anything. Okay, it's not. I would call what I just did, guys, a bad prompt. The reason for that is I didn't give it much context. I basically just said, hey, just do this. Let's try to give it a little bit more of an idea. Create a 2D model of the solar system with p5.js. I want it to be a top-down view. Let's see, an animated 2D model source. I'm going to be a top-down view and for the planets to be following their proper rotation. Oops. When I click, I want a shooting star to come out of my cursor. Make the— whoops, not bake— make the background stars that are slowly rotating to give the entire scene some life. Okay, let's try that. Make the colors awesome. Okay, all right, so this is a little bit better prompt. I've given it a little bit more of my vision, what I'm looking for. It's pretty cool because it knows what the solar system is. It knows, okay, yeah, these are the planets. You can see this, initialize planets, here's their relative orbital periods. You don't have to understand what this code is. This little stuff right here, this is called Markdown. It basically shows you or tells you what that part of the code does. You don't really need to know that, but a fun thing to know. But yeah, now we're much, much more complex here. So we're gonna copy this one, we'll go back into the editor again, we'll select all and delete this and we'll paste it again and we'll click play. Okay, it did not work. This is good, this is what we like. Okay, we got a reference error. So I'm gonna come right here and I'm just gonna copy this and I'm going to go back and I'm just gonna paste the error right here. Okay. Perfect. So now it's going to correct it really quick. All I did, I didn't give it any, any details. I just said, here's this. It understands what I mean because it's referring to the whole conversation. It knows everything that's happening inside of this conversation. We'll talk about that a little bit more in a little bit here. Okay, so guys, what I want you to do is, um, I'm going to go ahead and copy this. I'm going to come back in here again, but I want you guys to go ahead and dive in. You don't have to do a a solar system. You can do whatever the heck you want. What I am going to do though is I am— yeah, there's my solar system. Look how cool that is, right? I can click, I can shoot some stars. Pretty neat, right? It's kind of slow. Maybe I could tell it to speed up so it's just a little bit more fun. It could take a long time for— is that Pluto? I don't know if they included Pluto. Pluto's a planet in some circles, no pun intended there either. Yeah, it looks like it's Neptune. So anyway, okay, so Not very hard. I created something pretty cool. I could obviously keep going with it and talking to it and iterating it and say, okay, make it go faster, do this, make it more fun. But I think it's a pretty good one. I think it's a pretty good P5.js exercise here. What I'm going to do here, guys, is I'm going to copy the prompt that I did. I'm going to this one right here. I'm going to drop it into the chat in Zoom. You guys can copy and paste it. You can do the exact same thing that I did. Or you can say, hey, you know what, I want to build a little Pac-Man game, or I want to do whatever. You can build it and say, hey, yeah, let me use my arrows so that I can go around, build a snake game. This is a very low level in terms of the graphics available. You're not making Call of Duty with this one or GTA 6, but you're making very simple things. You could make a visualization of the world population from beginning of time to now on a graph. You can do whatever you want, just tell it that you want to use p5.js and click over into that p5.js editor. It should, it should provide a link. If it doesn't provide a link of where to do it, say, where can I do this? Right? And it will provide this p5.js web editor. Okay. Any questions?

Why, why did you use Groq versus ChatGPT for this? I mean, do you find that for like Java, for instance, like, does that, does that work better in Groq or like, like, I'm just interested in the choice and kind of the relative strengths.

Yeah, it's a great question. We'll, we'll dive in a little bit more to the difference in models. Groq is actually just, well, Groq 3 specifically is a fantastic coding model, right? As compared to like 4.0. 4.0 can do it. ChatGPT can do this exercise like no problem. It really can. But I find that Groq 3 is actually a better developer. And I only know that because I've read a bunch about it and that's what people say. That's what developers say and people that know say. And I only know that because I've read a bunch about it and that's what people say. That's what developers say and people that know say. The reality is I like using Grok a lot for research because it can search for like market research when I'm trying to figure out how do I sell this thing, right? Because it can search tweets, it can search all of X and everything that people are talking about in conversations. And so I use it a lot for that, but I like bouncing back and forth between the two. But if you want to use ChatGPT to do this exercise, you will succeed. Okay, good question. All right, this is a race. Whoever has their whatever they're building up first, go ahead and say, hey, I'm ready, and we'll have you share your screen. Meanwhile, I'm gonna make mine a little cooler.

I'm going.

You wanna— you have yours up?

Yeah, I think so.

Amazing. All right, we're gonna have you share your screen, Kelly. Kelly with a Y, go ahead.

Okay, share screen button. Hold on, I need to give access. Come on.

It's possible when you try to share your screen, it will ask you to restart Zoom.

Ah, it is. Okay.

And if that happens, just do it. You want to do it. You need to be able to share your screen. This is part of this exercise, guys. We're getting Zoom working for everyone. Go ahead and rejoin in a second. Update to mine. I broke it again. Who else has theirs ready? Go ahead, David, share your screen. And then Mike, it looks like you were unmuting to show what you got. We'll do you next.

Can you see it?

It's just started. Yep, here we go. Now, yeah, we— now we see— yeah, it's in the back.

So I just said create a map showing the location of the main archaeological sites in the Yucatán Peninsula. When I click the mouse, make the names of the sites pop up with 5JS. Cool. I did this and it shows me a little map. When I click on it, it shows me some of those sites.

So based on— I know it's not like a perfect map obviously here, but based on what you see, is that approximately like where they are? Like, did it get the locations right?

A little. So it's more south. Chichen Itza, yeah. Tulum is more towards the coast.

So try saying this. Say to Groq, say, hey, make the map look more real. And make sure the locations are more accurate. We got a fast typer here. I love that. Yeah, just click refresh. Sometimes that happens. Just click that. There you go. Yeah, regenerate. Just copy what you did and paste it again. I don't know why Groq does this sometimes. This happens like once out of every Maybe just refresh. For the whole page. Refresh the whole page. Yeah. Did it send? Did it send through Go Up? Go Up or did it not? Did it just completely ignore everything you did? Yeah, it did. Wait, hold on. Maybe— scroll up a little bit. It almost looks like it did. Oh, I see what it's telling you to do. It's telling you to actually upload a Yucatan map so that it can get it exactly right. That's cool. That's all right. All right. Well, you keep playing with it, keep talking to it, and that's cool. But good job. You generated something that's fantastic. Mike, go ahead.

I just edited mine, so it's still processing.

That's okay. Kelly, you ready now?

Yes.

Okay, uh, now I hit the share screen button and it should allow me to do that.

Yep, here we go.

All right, uh, let's see here.

A Yucatan map so that it can get it exactly right. That's cool. That's all right. All right. Well, you keep playing with it, keep talking to it, and that's cool. But good job. You generated something. That's fantastic. Mike, go ahead.

I just edited mine, so it's still processing.

That's okay. Kelly, you ready now? Yes.

Okay, uh, now I hit the share screen button and it should allow me to do that.

Yep, here we go.

All right, uh, Let's see here.

Hey, you have to make sure, Kelly, that you enter in or that you tell it to build with p5.js.

Ah, right, right, right. So hold on.

Just say it anywhere. Just say it at the end. Just say make this with p5.js.

Make this with p5.js like that?

Sure. Yeah.

Okay.

And send. Hey. You have to make sure, Kelly, that you enter in or that you tell it to build with p5.js.

Ah, right, right, right. So, ooh.

What are you in right now? AskAIChat.app? Just say it anywhere. Just say it at the end. Just say make this with p5.js.

Make this with p5.js like that?

Sure. Yeah.

Okay.

And send.

Oh, this was where it sent me for Groq.

Yeah, I think you clicked on the first—

says Grok Technology.

Yeah, I know. You found a wrapper. You found a fake Grok. It still will probably work, but—

do I need this one?

Yeah, that's where you want to be. That's okay. Go back to the other one. It should still work.

Sure.

And it's almost done. Yeah, you found what's called a wrapper, right?

This is fun.

What are you in right now? Ask AI chat.app.

Oh, this was where it sent me for Grok.

Yeah, I think you clicked on the take, like ChatGPT's technology or Grok's technology or something like that, and they, they wrap it and sell it to you separately. The first—

says Groq technology.

Yeah, I know. You found a wrapper. You found a fake Groq. It still will probably work, but—

do I need— I'll do this in a sec.

Yeah.

So, you can finish this one?

Yeah, that's where you want to be. That's okay. Go back to the other one. It should still work.

Sure.

And it's almost done. Yeah, you found what's called a wrapper, right? So, all right, yeah, it's still working. It's building. You asked it a fairly complex thing.

I did.

The gold glitter trimming, I think, is—

they take like ChatGPT's technology or Groq's technology or something like that, and they wrap it and sell it to you separately.

I'll do this in a sec.

Yeah.

So you can finish.

All right. Yeah.

There.

It's still working. It's building. You asked it a fairly complex thing.

I did.

The gold glitter trimming, I think, is—

I know that's going to— This is like a gem I've had in my mind already. So I was like, make it for me.

All right.

So interestingly, when I did it, it actually it displayed the preview right in Groq, which I don't know what.

That's okay. That's okay. Super cool. Yeah. If you do that inside of Groq, that's one way to do it. If you use X or something else, there's a couple programs that won't do that. I see. But that's another way to do it. Yeah, it works just fine. A lot of AI editors will allow you to make that happen. All right, Kelly, we're going to come back to you in a minute as soon as you have yours up and running. Who else has one ready to show? How about Kelly with an I? You got something cool?

I'm trying to do a pickleball game and I've rewritten it like 4 times.

I love it. But you're getting it playing and going back and forth or is it?

I'm trying to. I haven't gotten there yet, but I'm committed.

Keep working it and then show us. That's awesome. Connor, what about you? We've got our experienced developer here. I'm sure you've got something working.

Yeah, I'm working on my laptop, so it's a little bit harder to get around, but I have something I can share.

Yeah, please do.

Very simple. Um, can you see my screen? Yeah, it's just a basic Pong game. There's only one player. I can move left and right, has a ball and moves around and bounces off. That's about all I have right now.

So that's pretty cool. Um, you should tell it to create a second player that, that controls itself.

Yes, that's what I would move on to. I think I'm going to transition over to my desktop, it'll just make this faster.

Yeah, cool. Awesome. Thank you, Connor. It's fantastic. Amazing. All right, who has not shown their screen yet? Travis?

Yeah, here we go. All right, are you seeing my screen?

Yep.

Yeah, I'll rerun this. I asked it to show me the price of Bitcoin over time, separating the vertical line bars by month, and then make them interactive when I hover over them.

Oh, nice. You might have to click stop and then play.

Yeah, so here it is showing us by month. The prices, uh, don't look to be terribly accurate. They're like mostly just directional. You can see here We're about to, uh, fix about the moon for the first time. And then here we come up on 2023, '24.

Now it's just straight up forever.

Yeah, forever from here on out. Weekly buys, everybody. Um, what my price is by month That's cool.

I love that. Awesome, awesome. Very cool. You know what's cool about something like this, guys?

For the first time, and then here we come up on 2023, '24.

Now it's just straight up forever.

Yeah, forever from here on out. Weekly buys, everybody.

Eyes is— there's, you know, Now that you know how to do this, it takes you like, I don't know, 2 or 3 minutes, 5 minutes maybe if you want to work hard at it, right, to get something to, to show. You could take data from your company, you could take, you know, some presentation, some idea. Maybe you're trying to, um, what my price is by month. That's cool, I love that. Awesome. Awesome. Very cool. You know what's cool about something like this, guys, is there's, you know, now that you know how to do this, it takes you like, I don't know, 2 or 3 minutes, 5 minutes maybe, you know, pitch somebody on something, whatever, a sales pitch or a, you know, VC pitch, whatever it is, right? And you can now create something cool, right? A cool little visualization or animation related to if you want to work hard at it, right? To get something to, to, to show. You could take data from your company, you could take, you know, some presentation, some idea, maybe you're trying to, you know, pitch somebody on something, whatever, a sales pitch or a, you know, VC pitch, whatever it is, right? And you can now create something cool, right? A cool little visualization or animation related to whatever it is you're trying to do. I think it's pretty cool. Way better than just creating another slide on your PowerPoint. Whatever it is you're trying to do, I think it's pretty cool. Way better than just creating another slide on your PowerPoint. All right, Mike, we haven't seen yours yet, right?

Correct.

I just got off the screen, so I'm just going to say play it again.

Well, go ahead and share your screen if you—

if I can figure that out.

Find that button. Yeah. Yeah. Again, this exercise, guys, is a two-part exercise. All right, Mike, we haven't seen yours yet, right? Zoom and can we actually code something? Cool.

Anyway, I just coded like I just wanted a cartoon playing basketball.

Cool.

That's what it came up with. Interesting.

Interesting. Interesting shooting form.

Yeah, interesting shooting form.

I haven't corrected the shooting form yet. Typically we shoot, we shoot basketballs with our hands.

It's pretty fun.

All right, good stuff. That's a good one. That's a Hall of Fame one right there. All right, who has not shared? Anybody? I know Kelly's still working on it. Are you close?

Yeah, I jumped ship because it wasn't working. So I'm trying to do something more basic now.

All right, that's fine. All right, what about Kelly with a Y? You ready to get it working?

I'm just trying to figure out how to put it into the editor thing.

The code is done.

Share your screen and we'll walk you through it. This is good.

Share. This one. Yes. That's what I want.

Awesome. You're in the right spot. Go down to the code and see that little copy button right there?

Copy.

Hit that and scroll all the way to the bottom. You're there. You're there. No, you're good. Go back. It looks like you're already— go back to that p5.js other tab right there.

And then I— do you just hit play? And nothing happened.

Okay. And this is— you copied all that over here and then this is all you got was a gray box?

Yeah.

I'm gonna try again. Show me. Go back to Groq. Go back to Groq. I want to see your prompt again and the code that it gave you. Yeah. Scroll way up to the top. Okay. Scroll down. Keep going, keep going, keep going. I think you copied it too early. Copy it again. Go back to Groq. I want to see your prompt again and the code that it gave you. Yeah, scroll way up to the top. Okay, scroll down. Keep going, keep going, keep going. I think you copied it too early. Copy it again. I don't know. We're done writing. Yeah, go down again. Go down again. No, no, no, not that, not that, not that. Go down. Right there. Yeah, copy that code again.

This?

Nope. Wait, this? JavaScript. There you go. Yeah, and just push that copy button in the top right. There you go. Copy? Yep, perfect. We're all learning how to be really efficient guys here. All right, now select all.

Oh, there's kind of something in there.

Yeah, I don't know what the heck that is, but go ahead and select all.

I mean, it's a gym if you really wanted to make it a gym.

Again, this is to remind you again, this is not a— we haven't uploaded any. I feel like your code is more than 160 lines. Try pasting it again. No, it's the same thing. I bet it's the same. But yeah, you would have to provide 3D models of all of the different things you were talking about, Kelly. It did make your grass strip. So that's cool, right? It did. It's doing some fun stuff there. It's approximating things. It made a grass strip. It made a bunch of objects. It made 3D. It made it interactive so that you could do that.

And there's extra rooms.

So that's pretty cool.

Sure, yeah.

Awesome. Nice job. Nice job. OK. All right, next, Kelly with an I. I think last but not least. For sure.

So this is all I got. I lost my graphic, I lost my ball. So I don't know what happened.

Or with 3D, it made it interactive so that you could do that.

And there's extra rooms.

So that's pretty cool. When it doesn't work like I want it to work, I just say, hey, I can't see the ball, or hey, there's too much blue, or whatever it is. Just talk to it like you would talk to a developer. If you hired a developer to build you this game, what would you say to them? You'd say, I don't see the ball, this looks bad, try again, whatever. Go ahead and do that and we'll just watch for a second here. Bees add a pickleball that bounces over the net back and forth. So I was sitting one time at a— just in a group of people and had a friend telling me about a game that she wanted to create. This was a couple months ago, and she said, I want to create like a Flappy Bird, but instead of a bird, it's a bee, and there's flowers. And, you know, she's telling me all these ideas. And I'm just kind of like listening as I'm— I have my phone sitting there on the side and I'm just typing. As soon as she's done, I was like, kind of like this, and I just turned my phone to her and literally the game is auto-playing. It's like a, it's like a Flappy Bird game, whatever, that I had created just inside of ChatGPT using this exact strategy, right? Super easy. And she was just like, what the heck? Like, how is that even possible? How did you do that? Right? Kind of a fun, a fun party trick to be able to do. I don't know why it's—

oh, I don't know why it's so bad either. It's happening.

Yeah, Grok is struggling. It just said, I— okay, now it's— now it's working. Okay, I'm gonna visually 3D, clearly visible pickleball. Yep, that's good. I don't care why, I want you to just write some code. All right, I guess it's thinking out loud. Here we go, here we go. Now we're getting to the meat.

Okay, the first thing I asked it to do is have 4 players, and like I told it, like, player 1 to player 2, player 2 to player 3. But I think I just over— I think I got overexcited, right? Still don't have a ball.

Yeah, that's weird.

Anyway, I'm good if we just move on.

No, we can move on. We can move on. What we wanted to do here, guys, is just get this first element out of the way to where everybody knows how to share their screen, everybody knows how to copy code. Paste it back and forth, things like that. Now we can jump in and start to do some really cool stuff. I'm going to share my screen here in a second and we're going to dive into something just a little bit more complex than that. The beginning here, guys, was really about what I talked about. Hey, you know what? You've never developed code before, you've never written a line of code before, now you have. Now everybody has, at least with AI. You don't have to know what it said, you don't have to know what it meant, you just have to know that, hey, I asked for something, I brought it over here, and it still worked. If you took more time and if you simplified it, you took a step back, you could probably figure out some pretty cool stuff. I have 4 kids aged 7 to 12, almost 13, and all 4 of them can write code. They can write games using either p5.js or JavaScript or other things like that just because I've showed them how to do it with ChatGPT. They understand, oh, I ask it for something, I copy the code, I paste it into a web editor, and I click play. Then, oh, I want to change something and I tell it what I want it to do. All you have to be able to do is talk or type and you can do it. Mike, I see you've got a kid right there. Super fun to sit there and build a game with your son. Sometimes my kids will say, hey, can I play Roblox or can I play Fortnite? I'll be like, no, don't do that. They'll say, can I go build a game? I'll say yes. Really, really awesome to use your mind like that to build something fun and really make it come to life. Very cool. Let's dive into here now. We're going to do something new here. Close this out. Close this out. Close this out. And we're going to open up. All right. I'm going to, let me share my screen again. We have, so this is a folder, guys, that I'm going to give you at the end of the course. This has a bunch of different documents, overviews about, you know, how, how this works. Some of you have possibly seen a document or two kind of like this, but this is kind of some of the stuff that we'll be going over in terms of business ideas and validation and infrastructure setup and building an MVP. By the end of today, you will all be building really, really cool stuff. This is the schedule over the next few days. There's software and tools in here. You can see what are all the different pieces of software that Jon used in building these different things and how are they used for different purposes? How do we make that happen? Some of them are AI, some of them are not. What we're doing here in this course, guys, now the topic of this course, the title of it is How to Build a Business in 3 Days. With AI. And so we're going to do a lot of AI stuff, but we're also going to use a lot of non-AI things that you have to use to build a business. And so this is really more of an entrepreneurship plus the tools of how to do all that with AI driving every step of the process and making it all much, much, much faster. Okay. So, all right. One of the cool things that I get to do when I— I'll share my screen again in a second. But one of the cool things I get to do when I am in these courses is I get to build a new business every single week, and I come up with a brand new idea every single time. And, uh, and I'm excited to, excited to do that today. Um, let's see here, we'll share the screen again right here. Well, let's think of an idea. Let's think of an idea together. Um, let's see here. So I like a lot of things. I like Kayaking, and I've built some kayaking-related apps related to rivers and things like that. I'm a musician. I've got 3 pianos just in this room that I'm sitting, one right there, one right there, and one sitting over there against the wall. I like movie stuff. I haven't built any movie things, movie-related things or soundtrack-related things. Let's see here. What would be cool? Yeah, why not? We'll build something fun. So, I want to build an app— or yeah, a website. Let's say a website using React that allows people to discuss their favorite movies. We're going to work on this idea and figure out how to make money with it later. It's favorite movies, movies, movie music, actors, TV shows, TV shows, and more. That's all I'm going to say and we're going to click Start Building. This is Replit. If you haven't signed up for Replit, it's a good time maybe to do it while you're just watching the screen here on the side. Make sure you get all up and rolling. We're going to be using it pretty quick. We won't be diving into Replit quite yet for you guys. We'll come back to it in a minute. We're going to start in ChatGPT and do a bunch of other things. But right here, this allows me to quickly bring to life all kinds of cool things. All I've done is given it a simple sentence about my cool new Cinema Chat is what it's called. I hate the name. We'll I'll fix that later. But for now it's going to build something neat. Now right here, this is not just a fancy animation. This is actually writing code for me right now. And it's doing something called chain of thought reasoning. So I called it Cinema Chat up here, called it Cine— CineDisco. CineDisco.

CineDisco.

It's hard to say for me. I hate the name still. Okay, but it's starting to build it out. Um, so let's, let's, uh, let's pick on someone here. We're gonna pick on, uh, David. David, tell us how you get ready, how you get ready in the morning. Like, tell us your routine. How do you get ready?

I wake up early in the morning. I like to take some time to just, uh, read, meditate. Um, I like to see, check what I have to do, my agenda for the day. Then I jump in the shower, I get something to eat, and then just get going.

Awesome. All right, very cool. I didn't hear one thing. I hope, I hope you just kind of skipped over it, but I didn't hear brushing your teeth. Um, I want you to tell me though, David, tell, tell us how you brush your teeth in, in detail. What is, what's your, what's your routine?

So I take my toothbrush, I put some toothpaste on it, I, I put it under the— I get some water running start brushing my teeth, uh, up and down, left and right, going into the— what do you call it— the muelas.

The—

yeah, the molars. Yep.

Yeah, the molars. Um, make sure I, I go thoroughly, and then I rinse.

I don't know. And you're done. All right, I love it, I love it. That's awesome. So each of you, you know, probably thought about your own, you know, how do I get ready routine while David was talking. And, and the cool thing is— by the way, guys, look how cool this is. We've already got a company rolling. We've got at least an idea going. This is pretty cool. This is going to be fun to build. What David was doing just now was manually taking us through what's called chain of thought reasoning. AIs have not always been able to do chain of thought reasoning, but for about the last 8 months, they have been able to. Chain of thought reasoning means you take a task like getting ready and you break it down into smaller tasks, right? Get out of bed, say your prayers, exercise, meditate, eat breakfast, brush your teeth, you know, whatever it is that your routine is, right? It's going to break those down, but then it breaks down even further, right? Okay. Hey, my brushing my teeth exercise. Okay, I'm going to, I'm going to, I'm going to turn, I'm going to grab my toothbrush. I'm going to turn the water on. I'm going to turn it on. So it's maybe a little bit warm and cold, right? So I get the temperature just right and then I'm gonna get my toothbrush wet and then I'm going to open the toothpaste cap and then I'm going to put toothpaste on. I'm going to close the toothpaste cap and I'm going to put the toothpaste back. Right. It breaks down into this large task into subtasks, sub-subtasks, and sub-sub-subtasks all the way down until it figures out what it's doing. So that's what AI is now capable of doing. It can do very complex thinking with breaking down a problem If you say, hey, I need a new website, which is what I told it here. It thinks, okay, great. Let me get started on that. Well, a website probably needs some navigation, so let's create a nav menu. Based on this topic, probably want to talk about home, movies, TV shows, music, actors. Cool. We need some title. There's no images yet, so it hasn't figured that piece out. There's a couple images, there's no hero image. Probably want to join, maybe some discussions. And so it's just thinking what, what, what would this type of website probably include? And OK, now that I figured out kind of what I think it should include, now I'm going to start building it one little bit at a time. And the way that Replit works is it first will build a visual preview, right? Well, you can see here it's not interactive yet. None of the buttons work. If I click Create New Topic, nothing's going to work. And so right now it's creating a ton of different things. And so for those of you that are developers, you can see how much code it's writing really quickly. It's writing pretty dang good code and it's creating storage and routes and index. If you look over here on the left, you can open the files and you can see that it started to create a bunch of other things over here. You've got your client, your server, and your source, and your components, buttons and calendars, cards, carousels, charts, commands, dialogs. It's got all these different pieces here that it's pulling in. And building and writing fresh, right? Specifically for this app. And so it's writing thousands and thousands of lines of code right here, thinking about each step along the way using this chain of thought reasoning. And so most of the AIs that we're going to be using today, guys, are using advanced chain of thought reasoning now. And it's really, really cool. The difference between non-chain of thought or CoT and full-on CoT is really, really substantial. So while this, this, this will take probably another 15 minutes or so, we're going to let it just kind of go and we're going to bounce over to ChatGPT here in a minute. First, we're going to do a quick discussion about the business ideas, right? So I said to you in the email, some of you have business ideas, some of you, like maybe Wally cases, maybe you're not trying to build something new on the side for you, or maybe you are. Maybe you want to take this so you can learn both of those, but maybe you want to think of an internal project for Wally that you're like, hey, I really want to build out this new site or this new prototype or this new tool. Or this new whatever related to this, this new landing page, this new funnel, whatever you want to create, right? I want you guys to think of that. And so what we're going to do really quick is we're just going to go one by one, probably in about the same order. David, Mike, Kelly, Kelly with an I, Travis, Drew. We'll go through it and we'll have you guys just tell us what is your business idea. And if you don't have a business idea, say I'm not quite sure, and that's fine. So we'll start with David.

So yeah, I'm not quite sure. I have a couple, I have a couple ideas. Some have to do with tourism, some have to do— I spent some time in the US, so I saw Latinos taken advantage by the, this, there's their own people, you know, just because they didn't know the language and they just didn't know how to get places and proper insurance, things like that. So maybe just a tool that can help them find resources, you know, communicate effectively, things like that.

Very cool. I like that idea. I think it's pretty cool. I think it's very applicable. Obviously there's a lot of, a lot of discussion and noise going on between Mexico and the United States right now and people wanting to figure out how to do it right, you know? So that's pretty cool. If that's what you want to do, great. If you don't, guys, if you don't know what you want to do yet, guys, you can very quickly suss out an idea with ChatGPT. ChatGPT will help you dive in really quickly, be able to figure that out. Okay, cool. Mike, go ahead.

I mean, my first idea is just with chatting with ChatGPT, maybe too complex, but my first idea was I schedule my patients almost every evening because my schedule changes a lot almost every day and I try and optimize it for like the least amount of drive time. But I text, I sometimes do phone calls, and then I get halfway through the scheduling and someone says that they have an appointment and so I have to like redo the whole schedule. And so I was hoping to do something like the AI will send texts and schedule them. And then if they say 9 o'clock doesn't work, then it texts and asks a different time and reschedules it and optimizes for drive time because it can save me like an hour if I don't have to crisscross the valley.

Right.

And the tricky part is getting access to EMRs.

Right.

And ChatGPT, there's some other thing I can't even remember what it said, but it told me that I don't actually— anyway, ChatGPT was really helpful in like diving through that. But that's the idea, general idea.

I really, really like the idea. I think it's super cool. Not only— I, I think this is a tool you should build for yourself, and then you should set it up so that other people can buy it from you. Now that you can build that in the next 3 days, I think that's fantastic. Um, basically route optimization for anything. Right, right. And ChatGPT does a pretty good job of that already, honestly. The big— a lot—

the auto scheduling would be huge, and I don't know what's even capable.

Anything you can imagine is possible. We're going to be talking a lot about, guys, things called APIs, right? An API is basically where you're talking to another service. So you're going to connect to an email service, a texting service, Google Maps API. Probably ChatGPT or some other AI to do some of the analysis and things like that. But you'll be able to have it do everything that you want. Right? You just are gonna tell it exactly what you want and we'll, we'll help you build. It'll be awesome. Alright, Kelly with an I. Oh, man.

I have so many ideas. That's my problem. For Wally, I am in charge of sales channels, so I want to create a dashboard where I can get all of my sales channel information in one place. Right now, we just don't have a place that it's all there.

So where, where does it live right now?

Huh?

Where does all that sales information live right now? Is it in Slack?

Um, some of it's in Shopify, but some of it doesn't pull in from Amazon. Some of it's in Etsy that doesn't pull all together. So there are tools. I just, I need it all in one place because we're growing so fast. So that's my goal for this. I also really want, um, I run a nonprofit foundation, um, that helps families that have kids with heart defects. So I also want to start like a Reddit for heart parents where, like, right now all of the support is in Facebook, basically. But I want to get it out of Facebook because I hate having to search through Facebook through things. But I want a place where the heart parents can come, um, yeah, kind of like a forum where they can discuss and share everything, right?

That's amazing. For the purposes of this course, you could choose either one. I can tell you that the Wall-E one is a little bit more complex And so it might be good to, to try that in the course because building that nonprofit one, you'll be able to do in an hour. By the time you're done with the course, you'll be able to do it in an hour. It'll be super, super easy and you'll be, you'll be all enabled and ready to rock. So, so that's awesome. And guys, I don't want you to feel like, oh my gosh, if I choose the wrong business, like my life is over. I've had so many students build 2 or 3 apps during the course because they'll just be like, oh, you know what? I'm going to catch up and switch my direction. I'm gonna do this and I'm going to go this direction. I had one guy launch his ads by the middle of day 2, right? And he had, he has customers. He's made $6,000. He started about a week ago, right? He was the last course that I did. So actually a little more than a week ago.

Okay. I have one more idea. I just want you to tell me how hard this is going to be.

Oh, go ahead.

Yeah.

So this is, this is personal. So I have kids in like 8 different things, right? And I have all these apps for each of my kids' things. And I want, I want to roll it up. Like I want to be able to log in, see all my notifications. I want to schedule for each kid. I want the calendars from all those apps. We're talking like WhatsApp and, and TeamSnap and Band. Like, I want it in one spot.

Yeah, that's a little bit difficult only because I don't know if things like Team— I, I have 4 kids too, right? And we're running, they're all over the place. And I actually started building this solution in my last class. I started messing with it and I didn't finish it. And, and you should absolutely do it, but I just don't know what kind of information we can pull out of TeamSnap. Like, is there an API? Is there a way to send a message from there? I'm not really sure. Can you forward it to a phone number or something? I don't know. It's a fun problem to think through and I'm sure it can be done, but I just haven't asked all the questions. All of it is possible. All 3 of those ideas that you've talked about are most certainly doable by you at the end of this course, right? Very easy. So yeah, pick one, pick one and run with it. Drew.

I have about 20 different ideas, so it's a little tricky. I think one that could be helpful to get a jumpstart on here is a buddy, a buddy and I have talked about starting a new company. And one of the, one of the key components of this company is essentially allowing people to interact with their business systems in a more cohesive way. You know, so that could start with a simple chatbot, but it's pulling data from maybe 5 or 6 different business systems. So maybe it's pulling from your Slack and your Google Workspace. And so if I want to know, you know, no information that would require 2 different information systems coming together. And then we want to build some stuff on top of that. But that's kind of the basic, like, because I think just even creating a shell where someone could log in using Google, and be able to start plugging in their various business systems to interact with it, that, that could be a good start.

Yeah, that's cool. Um, I love it. I think that's great. Um, all right, Kelly with a Y, I think you already told us, I think you're going to be building a website for your, um, whatever you do, you know, helping the bodybuilders build their bodies.

Yeah, but now that I— you already put us in a Grok, but I'm going to make it as an interactive gym where people can select, because here's the thing is I have 7 different businesses in one. So if they want posing coaching, they go into the posing room, it gives them package options. If it's training, they can go into like in the front desk aspect, we'll tell them where to go on the website too.

That's interesting. Yeah, it's, it's difficult. I, I'm building a site right now for me personally that's just called joncheney.com, right? I've had my website up for over, I don't know, maybe 13 or 14 years, and if you go to it, all you'll see is my piano music. I write music and I sell the sheet music on there, and that's what that website is But I do way more than that. And so if someone goes to my LinkedIn and then they land on joncheney.com, they'll be like, oh, this is just music, right? Like, I want to show everything that I do. Very similar to what you're talking about, Kelly. I obviously do entrepreneurship and coaching and consulting and AI training. I have these different businesses that run. I do all these things and I like to write a lot and I go on adventures. Anyway, I do lots of things. I hurt myself a lot. Anyway, so I want a place to put all that, but it's hard to have kind of an overarching site that then has 7 mini sites inside of it. Um, it's a fun problem. I think you should try it. Right. But I'm interested in how you tackle it. And if you figure something out really nice, I want to see what you do. Cause I'm actively working on how do I make navigation within navigation to not confuse the user? Cause if I, if I have a piano, this is, this is important for everybody, guys. Right. If I have, if you have someone who's coming to you for piano music and then you show them entrepreneurship stuff right away on the site. They might be confused a little bit, right? And so you have to figure out how to funnel the people to the right spot or to very quickly let them land and be like, oh, this is Jon's like overarching site. And boom, there's my music place. That's— I want— I care about Jon's music. I don't care about his other stuff. Or I care about his entrepreneurship stuff. I don't care about his music. You know, you have to be able to let people find where they want to go without confusing them too much. So that'll be your struggle, Kelly. And Yeah, I look forward to seeing how you figure that out. And you'll work with, you're gonna work with ChatGPT to figure it out. It's gonna be awesome. Jorge and then Travis.

Yeah.

So I've been wanting to, there's a few ideas. The one that I've been wanting to do is in essence, like put, I would, I don't wanna say a bot, but an AI that helps transactional sales. So anything that's like $99 less, like less than $99.

Okay.

Where interacts with the user. Where that's on the website, through email, or through text by like an identifier. So if they can identify like at the very beginning, they interact, like it captures the phone number, the email, and then, um, uh, the chat.

And so basically I can text a phone number and I can get all the way through checkout without ever leaving the text.

Yes. Or like, or even like if you're, if you start through, like maybe you start through email. And then all of a sudden you continue with text, it knows the conversation you had through email. It's not starting all over again. It kind of picks up where they're at.

Very cool.

No idea how, like, how complex that would be, but that's—

No, I've already, I already kind of built it in my head. That's fantastic. I love it. It absolutely works. Okay. You're going to be doing APIs with Twilio and with SendGrid and with ChatGPT probably. With OpenAI, not ChatGPT, with OpenAI and connecting those all together and building a system. It's, it's really cool. And the, the, the proper name for that is a headless transactional system, right? Headless means there's no website. There's no, you know, you're just doing it through text or through email or whatever. You can do it anywhere. It's headless. It just kind of lives. That, by the way, guys, Jorge is very smart. That is the future. That is how you will interact with so many companies in the future. There are a lot. It's very cutting edge, But so many companies are working towards that. Wally, you should be doing that. You should be thinking about how do we, how do we interact with people without them even having to land on our website, right? How can they land? How can they finish their purchase on social by the comments or the DMs, right? That's the type of stuff that Jorge is building there. Very cool.

What did you call it? What's the right word for it? I never have a word for it.

I just kind of headless, headless, like without a head. Yeah. Sin cabeza. Um, right, Travis, go ahead. And then Connor, you're not— your video's on, so I forgot about you.

But yeah, I'm thinking, uh, I'm thinking an internal tool for Boostly, something that does like, uh, early churn warning system. We don't have anything like that right now. So we have— I think there are probably ways that AI could help me find some common, um, signals that we get from customers before they churn that I might not know about right now. And I want to know what those are, and then I want to Maybe create different like levels of warning, you know, have like a red flag and a yellow flag list or something like that for the people who are most at risk for churning in the near future. And then maybe send some notifications to my customer success team to let them know that people in their book are at risk of churning and so they can get ahead of it.

What kind of data do you think you'd be looking at and where does that data live right now?

Yeah, luckily we are, uh, all of that lives in a, uh, pretty massive Google Sheet that gets updated monthly with the most recent billing data. So I should be able to plug into that. Um, we're in the process of moving that out, but it won't be— we're still weeks or potentially months away from that. So I think this will help be— this will be like an initial build that I think could inform what we do in the future.

Very cool. You'll be able to, you'll be able to, you know, if you build it right, which, which you just have to tell it to do it this way, is say, hey, we might switch out this Google Sheet for something else in the near future. So let's make sure this piece is hot-swappable, right? And then it'll be just, yep, switch the switch where it's pointing and it will, it'll figure it out. There might be a few variables you have to change and things like that, but, but it'll be an hour fix to update it once you move it to a new system. So that's cool. All right, Connor.

Yeah, sorry, I'll get a webcam for my desktop.

But you're good. But okay. Yeah. Yeah.

So with Wally, I work for a lot of what we want to do is centered around data like Kelly was mentioning. On my end, it's a lot of just sales data, organizing that into one space and figuring out a tool to forecast our inventory sales. And then also using things for development on Shopify side as far as theme development or functional development there as well.

So, okay, very cool. All right, guys, this is awesome. We're about an hour in, a little, little over that, and we've accomplished— we're kind of right on schedule actually to do everything that we need to do here. By the way, guys, if you do ever need to go to the bathroom or get up for a second, jump on a quick call, feel free to do so. I'm going to keep going. All right. I'm going to go for 3 hours straight. I am ready. I've got my water. I have to go to the bathroom. I'll give you guys an assignment and I'll go really quick and I'll come back. But if you want to eat or drink during this, this is totally fine. But we're just going to crank and go, go straight through. Okay. All right. Now it's time to dive in. We've kind of done all the preliminary stuff. You guys have seen that AI can code. Let's check on our— let's check on our Synodiscus. It's starting to— Starting to, you know, join the community. It looks like a button now, kind of. Create new topic. Create a discussion. Post discussion. Is anybody else impressed that this thing, like, you know, kind of works now? Now this page has not been built yet, it looks like, which is fine. The home works. Some of these things work. Popular topics. That's pretty cool. I don't know. I'm always impressed when it comes up with these things. This is the first step. This is where we start. Now we start to build out from there. Now, here's the problem. This is a terrible way to build with AI, a single prompt. It's a sure way to fail because you got to tell it, otherwise it is going to use its beautiful, wonderful chain of thought thinking and it's going to think for you. I guarantee you its chain of thought thinking is not the same as yours. You have to guide it. You have to direct it. You have to tell it how to, you know, what you're trying to accomplish and why you're trying to accomplish it and how you're trying to accomplish it, right? So we are going to use ChatGPT to do that. This, I love ChatGPT, guys. Obviously, there's a lot of reasons why. Where on earth? My library? What on earth? Oh, there we go. Like, what the heck? Where's all my stuff? All right. These are all my cool pictures I've made. All right. So we are going to be using inside of ChatGPT, guys, something called projects, and we're going to go in there in a minute. What I want to do first is we're going to talk a little bit about context and how it relates to AI. So context is— there's several layers of context and several— there's a couple meanings of context as well. So we'll cover all of them. Basically, I mean, context, just like the word itself means, is kind of understanding what you mean. What are all the details around that request? What are all the details around what you're talking about? Hey, I have no context. You just said something. I don't understand what you're talking about. What are you talking about? Oh, okay, I got it now. Our brains need to understand context before we can start processing correctly. We have pretty awesome quantum computers in our brain, in our heads, but they still need context to figure out exactly what's going on. ChatGPT has a great, what's called, memory. Okay, by the way, for those of you that are dealing with, um, uh, I don't know, call them trade secrets, maybe you're worried about the world knowing about how you do certain things. If you're worried about that, um, and you think that ChatGPT is going to steal your data, you can come in here and you can say improve the model for everyone and turn that off. Okay, and so I keep mine on because I don't really think ChatGPT is going to steal the things I talk about. I think it's just learning how to talk to me and how to talk to other people based on the conversations it has with me. But if you had, you know, sensitive data, things like that, maybe you could turn it off. I personally would put my Social Security number into ChatGPT, just FYI, because nobody's going in and reading those. They're all encrypted, the messages. No one can go in and do that. It aggregates data and puts it all together, but it doesn't create— it doesn't— it's not going to say, oh yeah, what's Jon Cheney's Social Security number? Here it is, right? It's not gonna do that. So anyway, that's just a little sidebar. Back to the memory. Right here under personalization, you'll see this little memory. There's this little kind of button right here, and I'm in the browser-based one here, manage memories. So memories basically helps it understand who you are, right? It knows what I am. Hey, I'm fluent in Spanish and Chinese, and I'm a whitewater kayaker, I'm a member of the church, and I'm an entrepreneur, composer, consultant, right? It understands things that I do because I've talked to it about it, right? It even understands some of the last, you know, latest— hey, build sites with Replit, scaling GenAI API, it does all this, right? So it understands different things that I do. So that's part of its context. When it answers a question, it will consider the things in its memory. And if I— so if I say, hey, you know, I need to build this new site, it's like, great, hey, you like to use Replit, here's how you can do that for this, right? It's, it just It knows that without me even having to tell it, right? Or if I start talking about kayaking, it knows I'm an advanced kayaker. Okay, yeah, hey, you know, let's take this to a higher level. You know what you're talking about. So that's pretty cool. A second level of— so that's the first level. You've got the memory, the entire system knows who you are. If it doesn't know who you are, that's okay. It's okay. It'll learn over time. Number 2, okay, is the conversation itself. Some people erroneously think that ChatGPT knows everything you've ever said to it. That is simply not the case. The AI is not capable of that. It's too much context, meaning you're overloading its context limit and it can't understand everything. But if I come in here and I say, all right, I want to talk about this Mother's Day note. I was writing a Mother's Day note tongue-in-cheek for my mom, proving that I could write a note for her that was really nice. With just one tiny, you know, one tiny little phrase. Anyway, she's like, I still like it. Anyway, so I could come back to this conversation and just say, you know, make the note shorter, right? It immediately just picks up where I left off and it understands that context. I don't have to give it any more context. It knows who I am. It knows what's going on in this conversation. Great. Here is the, you know, here's the The shorter, the shorter note. Okay, so that's the second level of context, the conversation itself. You've got first level is that the third, and one of the reasons why I like ChatGPT more than other platforms for organizing projects is the projects tool. No other, no other AI out there has a tool that is as robust as this. And it's really not that even complicated. I'm surprised that other, other AIs have not taken the time to build out this feature.. But here we're gonna call this, you know, movie review site or movie discussion site. I don't know what I'm gonna call it. We're gonna call it movie discussion site for now. Create project. Okay, so this is my business idea. What you call it, what you call your project does not matter. So don't, don't fret.

Okay?

So we're gonna build a project out and then what you can do inside of this project to give it more context to understand what you're talking about, you can add files to it, right? So maybe for WALL-E, you have, you know, SOPs or you have documents, you have sales things or whatever you have, right? You could drop all that stuff right in here and it'll just kind of start to understand what you're doing. Maybe you can, in a second, you'll be able to tell it, hey, this is our website, this is what we do, whatever. And you can just kind of get it to start to come to life on that. But you can drop 100 files in here, 2 files in here, whatever it is that you want that might be related to it. I don't have any files related to my movie discussion site because I've never thought of this idea until right now. And so I don't need to add anything and that's fine. Okay. Now this instructions is actually pretty important. This is how the ChatGPT or how the GPT is going to respond. It's going to talk to you like maybe you want to speak in Spanish. That's fine. It can do that, right? Maybe you want it to reference some file that you gave it. Maybe you want it to act in a certain way. Okay. So I'm going to type in a few, a few ideas here. So we're going to, we'll do a couple of different types. We'll say whenever you address me, Call me your highness. Okay, and we're going to say, um, I want you to act like Steve Jobs and help me build a beautiful website that will have a killer user experience. Give me ideas to build better and more efficient and to think different. Apple's slogan. All right. I just randomly thought of that. Then we're going to say this, don't agree. Come on, fingers. We got this. Don't agree with me just to agree with me. Actively disagree with me and actively seek to improve my ideas. If I have a bad idea, tell me. Really help me get to the most optimized solution in everything, everything we discuss. Okay. All right. I'm going to click save on this and then I'm going to share this prompt with you. You obviously can modify it. You don't need it to call you your highness. But if I then tell it, let's give it the same prompt. Let's go like this and we'll go like this and we'll say we're just going to do that. And all right, so now I'm going to give it my first prompt here and it says, your highness, the idea of a movie discussion site has potential, but only if it thinks different. Okay, awesome. So now I've got Steve Jobs coaching me to build a cool business. And so I've given it just a high-level description, not a lot of details. We're going to dive into that in a second. Okay, when you guys dive in, you're going to do this in a minute. You're going to build your project. Maybe you're already building your project, but you're gonna, you're gonna copy, you know, you want, you do want to put the most important thing that you put in there, guys, is the don't just agree with me on everything. ChatGPT will make you think you're the smartest person in the world, and you're not. Neither am I, right? We don't know everything about everything, but it's going to make you think like, oh yeah, you're just thinking the best. This is so awesome, right? No, actively disagree with me, help me get better, right? You need that third paragraph. That's really important. If you want it to coach you like Steve Jobs or to coach you like somebody else, maybe you have somebody else you want to coach you, right? Kelly with a Y. Maybe you have a bodybuilder that's really famous. Maybe you wanna be coached by Arnold Schwarzenegger in a fun accent. But you can ask anybody you want to coach you and train you and it will be able to pull from all of the different, you know, books and everything that person has ever written or said, right? You can have them, you can have them do that. Okay, so I'm going to give it this idea and then it's going to, you know, it gave me some ideas. I'm kind of going to ignore what it said here. Um, so what I'm going to do is this now. I'm going to start, we're going to flip the script. This is one of the first cool things I'm going to teach you guys with AI. We're going to flip the script. Instead of us asking a bunch of questions, we're going to have it, we're going to have ChatGPT interview us about our business idea. Okay. And it's really, really easy. So we're going to say, interview me one question at a time about my business. Ask me about the branding, the name, the USP, the monetization strategy, the target audience, and everything else No, not everything. Let's say anything. Anything else you believe is important to discuss in order to build branding. I want to actually do this, the vision. All right, and let's see, important to build a great business plan. Okay. All right. So now I'm going to go like this. I'm also going to drop this into the chat so you guys can copy my prompt as well. And here we go. Here's my question 1. What is the deep emotional reason you want to build this movie discussion platform? So this is where we part ways. So guys, for the next 10 minutes, it's 10:24 right now. We're going to go to 10:34, at least my time here in Utah.

Okay.

Um, we're going to go for the next 10 minutes and you guys are going to build a project. You're going to, you know, add the instructions to it. You're going to tell it your idea. All right, and then you're going to ask it to interview you. Okay, one question at a time. I want you to get through about 8 questions in the next 10 minutes. So it's now 10:25, you get an extra minute. Okay, 10:35, we're back. Okay, so I'm going to keep typing, I'm going to keep sharing my screen, and I'm going to type and you guys can watch and look if you want, but you should be working in ChatGPT on your own thing. And feel free to ask questions as you go if you don't know what's going on or if you don't know how to get to this spot. By the way, guys, when you— when the AI asks you a question, if you don't know, like, like maybe the name of your company, you can say, I'm not sure, give me some ideas, and it will just talk to you, right? You don't feel like you have to question, answer, question, answer. You can discuss it, and as soon as you say, okay, this is what I'm going with, it'll move to the next question. Hey, Jon.

Yeah, I'm just going to show my lack of computer abilities here. You're good. I'm trying to get my ChatGPT.

I don't know if I can share my screen with you somehow. Go ahead and do it.

Um, I just can't get like this actual window. I know it's something simple, and normally I used my wife to help me, but she disappeared somewhere.

You're good.

So I'm just trying to like— I don't even know how to pull this over. Like, I'm missing— I don't know. I mean, I have two windows in my Google.

Yeah. So it looks like I have the tab, the left tab open. You know, you're fine. What are you trying to do exactly?

I'm just trying to, like, be able to see this. I can't pull it over.

I don't know. Is it the project that you're trying to find? This is the ChatGPT.

I was going to pull up the actual project.

Yeah. Yeah. So up there in the top left, you see that little— there's a little button. Yeah. Top left of the ChatGPT screen. Yeah. There's a little blue dot right next to it. See that? Oh, yeah. Yeah. Yeah. Click that. Yeah. There's your sidebar. So then this is where you would have your projects. Can you—

I'm trying to pull this in. Do you know what I'm saying?

Like, you see how it's not like on a full screen? Like, I don't— I usually can pull it over. Yeah, you should be able to just grab. Yeah, yeah, you should be able to grab right there.

I can do that one. I know, anyway.

Yeah, close out of that window right there or that little tab right there.

I need to move this bar.

There we are.

So I can't even close it because I can't, I guess I can close it here.

There you go.

There we go. Now I can finally pull it over.

Thank you. Yeah, yeah, you should be good. Yeah, perfect. Yeah, keep, keep going. Yeah, jump into that project and get that thing rolling. Okay, thanks. Okay, here, I'll stop sharing for you. There you go. You guys, I know you might not be completely done yet, but take another 2 minutes. We'll give you a bonus 1 minute, maybe another 90 seconds or so, and we'll keep going. You should have gotten through probably somewhere between 6 to 10 questions. Should have a name for your company. You should have some ideas about how you're making money. If it hasn't asked you, you could say, hey, let's, let's discuss monetization, you know, or whatever, you know, if that's, if that's part of what you're needing to do. If you're building an internal tool, then you might not need to cover that, but just make sure you talk about all the different things that, that you want. And obviously in your, in your own time, you can dive deeper in this type of a thing. I've spent 2 hours in an interview like this, you know, talking to ChatGPT, really diving deep into things. All right, guys, let's go ahead and dive back in here. So I gave you guys a couple extra minutes. I know you might still be working on it, and that's fine. Go ahead and keep working on it while I talk here. The next thing that we're going to do, I'm going to share my screen. Next thing we're going to do here, you should have gotten through some you know, some good— let's see here. Yeah, that's right. So what we're going to do next is we're going to, we're going to build a business plan with all of these questions. So it's asked us a bunch of business questions about, you know, what we're doing. Come up with some cool ideas. Here's what you want to work on. And, you know, again, I'm working on this movie site and trying it right now. I'm working on the name, and I kind of kept that to last. Normally I start with that first. And when I'm, when I'm going through the name, there's a few things that I'm going to think about, especially if I'm building a new site. You know, let's say I like Luma, right? Luma is kind of cool. I'm going to check really quick. I'm going to go to luma.com. All right, I'm going to see if that's available or if it's already something. It looks like it's available, so I might go to GoDaddy. I'm sure it's not available. It's too good of a domain. Not Luma, not Luna. Yeah, Luma's available, or Luma— yeah, Luma.com is not available. I can go down, I can look at a couple other things. Luma, love Luma. Luma NYC. Luma's a pretty dang good— is a pretty dang good thing here. Because it's 4 letters, it's, it's taken, right? Luma. And I think luma.io is like a, isn't that a, I swear Luma is something. Anyway, that's fine. But I'm always going to check here. There's a couple of different tools you can use to check if all the socials are available. I might bounce over to Instagram really quick and just do a quick search. Is Luma available? I'm going to, I'm going to do a little bit of brand checking before I kind of lock it in. Obviously the name of something isn't the end-all, but for me, you know, it's, it's something that, um, I don't know, something that really, um, drives a lot of things for me personally. Um, yeah, Cinemark, already a theater, theater chain. That's a terrible idea. So I've asked it for a bunch of ideas on name ideas. I got to come up with something. I'm trying to just grab one that makes me Cinevia. Cinevia. Cinevia. Cinevia. Cinevia. I kind of read Cinevia, even though I understand it's Cinevia, like away, like the via. We'll try it, we'll try it, we'll just roll with that. I'm just gonna say Cinevia. Flicker. Kind of like Flickr. Is Flickr something? I'm sure Flickr's already something. Kind of— it's kind of more playful. Like, you're the best place to be a photographer online, and it's already spelled wrong. Yeah, whatever. I'm not gonna worry about it too much because it's not a real company. So let's roll with Sinevia. Oh, whoa, that did not listen to my brain. Um, please create— guys, this is something that you're going to want to do as well at some point— please create some brand colors and fonts that will help us accomplish what we're trying to do here. Okay, so now I've asked it to create a bunch of different things, right? You've got, you've got names, you've got hopefully some, you know, USP. We haven't gone— I haven't gone over monetization strategy yet. I'm probably going to ask it really quickly to just do that. Please come up with a good monetization strategy. Typography, it's using some of those.

Enter.

I actually like that one. That's a good one. And by the way, guys, if you're building your own business, like, I know I'm kind of just going fast and cranking through, but you're probably going to sit here, you're going to read, like, deeps into black, real gold. Old amber rose. Like, do I like these colors? These are kind of cool. I like them. They should work pretty well. I told it to make a cinematic experience. Okay, then I'm gonna have it come up with a good monetization strategy really quickly just so we have a full business plan. Maybe you already know your bit— your monetization strategy, right? Maybe that— maybe that's not important, but you want to tell it. You want to say, hey, right now this is how I charge. This is how I make money. I make money when I go visit people at their houses and I, and I work on clients and I work on this, right? This is how I make money. Tell it what you're doing, even if it's just at a very, very high level. 2 or 3 sentences is all you need. Maybe even just a few bullet points can get it done. Okay, so, um, I don't know exactly what it's trying to do here. Premium membership, creator community-led monetization, affiliate revenue streams, limit 50 for free users. Yeah, anyway, it sounds like we're going to make something that's like, hey, here are my favorite scenes and here's soundtracks and things in movies. That's kind of how it's going to build. Okay. So now it kind of has an idea. I've gone through, I answered probably 12, 12 questions here. Like I said, you probably are starting to enjoy this if you have a real idea. Like you're probably starting to be like, man, this is like starting to turn into a plan. But now we're going to do this. Please create a 1,500-word business plan using the Canvas tool based on these responses or based on this discussion. We'll say that. Based on this discussion. Copy. Um, and I'm going to drop that here into the chat so you guys can do the same thing. But I'll show you what happens. We're using a new tool inside of ChatGPT called Canvas. What that's going to do is it's going to open up this nice new thing here, and it is going to create a business plan. The Canvas tool inside of ChatGPT is basically like a Google Doc or a Word document, right? It's editable. You can change it, whatever you can. It lives as a document in the conversation itself, right? So you can see this little document right here. It's being, you know, instead of answering in the chat, it's creating it using Canvas and it's kind of telling us here's all the different things that it's going to do, right? And so here's the architecture, here's some monetization strategy, etc. That's pretty fun.

Okay. Advantage of creating within a Canvas instead of just keeping it in the chat.

Because it keeps it organized and you're going to refer back to it several times. Yeah, this is a— I like using it in the canvas. And if it didn't write it, if it just answered in the chat, just say, hey, please throw that in the canvas really quick. And it'll just, it'll just put it up and it'll create a document. But you want to save this because we're going to make this plan, then we're going to make another plan. We're going to bounce this back and forth on some different things. So yeah, anyway, I came up with a tagline. If movie nights raise you, you belong here. I think that's pretty cool. Like, if you just love movies, like, yeah, come on in. Okay, so that's pretty fun. Um, so once you have this business plan, I'll give you guys— I'm going to give you guys 60 seconds to build this business plan, and once you have it, then we're going to dive back into Grok for some market research. Okay, I'm going to close a few windows while you guys figure out how to catch up there. Okay, I have too many things. Okay. All right, guys. So you've got your business plan. Now we're going to jump into— we're going to jump back into Groq. So here's what we're going to do. We're going to copy your business plan. There's a little button right here. You just click copy. And that's now copied this beautiful business plan. And we're going to bounce into Groq. You're going to start a new conversation, okay? And you could watch me and then you can do this. All right. All we're going to do is we're going to say here, please review this idea and give me feedback. Whoops. On the market sentiment around this plan. Give me ideas on how to improve plan and focus in on things that are working well. Okay, then I'm going to click deep search and then I'm going to hold Shift. When I hold Shift and I click Enter or Return a couple times, then it won't send the message to the AI. Okay, if I hold Shift, I can enter, enter and do this. I'm just going to paste my business plan. And I'm going to click here first. I'm going to go like this. I'm going to copy this for you guys. I'm going to click send. Okay, I will drop this prompt in the chat. So this is in Groq, right? Please review this idea. And I had it use the deep search function. The deep search function is going to take, you know, maybe 90 seconds, maybe a few minutes, and it's going to search 100 different websites out there. It's going to check, you know, Twitter. It's going to look at the market sentiment. What's going on? Is this a good reason to start a business. It's going to check on what you're doing and give you some really good feedback. The reason—

I did have a question. Why did you use Grok for this instead of ChatGPT?

Because ChatGPT does not have access to Twitter. That's the simple reason. Twitter has billions and billions of tweets every day of people talking about every topic you can imagine. If you want to know what people think about wallets, right, or think about whatever your topic is, ask Grok. Grok can tell you, can do deep search, and it can search for all of this different stuff. It says searching for demand right now. It's going to check websites and it's going to check Grok. It's going to check all kinds of different places for— not Grok, it's going to check X for all these different market sentiments that are out there live right now. So that's why I like it. Yeah, see, analyzing X posts. That's what it's doing right now on this step. It's awesome. And you can ask it specifically to do this. Hey, please check X, make sure that, you know, tell me what people are saying about this, this thing. And then when it's done, it's going to give you some key points and some feedback. And I recommend you read it right now. You might not be able to read it all right now because I'm going to be moving pretty fast. But take 60 seconds, guys, and get through your deep search.

I'm still not there, Jon. I'm still kind of putting on, on ChatGPT, but what was the prompt that you put on there?

Yeah, I dropped it. I dropped it into the chat on Zoom. Got it. Yeah. Okay. Yeah. Any, any prompts that I'm using, guys, I'll always drop on the chat. Yeah. Take a second, guys. I know this is a part, a place where I kind of rush you a little bit because you want to get it right. You know, it's got so many great questions. You want to develop your idea out. So I'll give you guys a couple minutes to catch up to this spot. Take a minute, you know, if you're waiting to, to read. And I'm going to eat, I don't know, something here in my box. I went on a no-sugar 30 days in January, and I found so many— first of all, I felt amazing. I highly recommend it. It's so hard, guys. It's so hard to like really do no sugar, no, um, no unnatural sugar, no added sugar. So I'll do, you know, fruit and, you know, like this That's It bar or whatever, no added sugar. But I found a lot of really good stuff. I feel way better, probably lost like 7 pounds. I'm gonna eat this while you guys catch up for just a minute here. I am going to read what Grocker's saying to me. And this idea for Snevia is pretty dang good. Cinevia. Cinevia. I don't even know how to say it, but There's definitely some opportunities here, but man, this is a full-time job. This is a full-time gig. But I feel like I could build this really well. It would be really fun. You basically just post on social media every day, I don't know, 12 times a day, like a huge amount. And like Instagram specifically, I think would be really good. Facebook probably too, and probably Twitter, honestly, probably X. Just post movie scenes, right? And then, but then with your like little analysis of, of like, here's, this is how it makes me feel, this is why this is my favorite thing. And then you get other people putting their own, their own comments and thoughts and their own things, and you just build up a community and then provide a place for people to go and hang out. Yeah, this could, this could be a multi-million dollar business in probably, probably a year. All right, guys. So for those of you that are caught up, great. For those of you who are still catching up, keep working it. Well, I'm going to go a little slow to just kind of let you catch up for the next 9 minutes or so. But after you've done this deep search thing, you can actually just click right here and switch it to think instead of deep search. It's actually going to change the model. And change the amount of chain of thought reasoning it's doing with all of this. Okay. And then I'm just going to ask it to think. I'm going to say, consider ways which we can improve this plan based on the feedback. Let you know. Okay. So I will copy this up. And so just drop that into the Zoom for everybody. This is the think. The think prompt for Grok. So now we're just going to ask it to give even more ideas really quick. And by the way, if you don't have Grok, guys, and if this isn't working, you can obviously do this in ChatGPT, but you would do it inside of a different platform. You would— or sorry, a new chat, and you would probably switch to one of its search models, right, where it is searching the internet, things like that. ChatGPT has several ways to make that happen. You want to go between different models. You want to use this. The idea here is to use multiple models to get feedback on what you're doing and to bounce back and forth and to refine back and forth. In a second here, we're going to take everything that Grok has said and we're going to come back into ChatGPT and we're going to refine our business plan even better. Right now, this is giving us a really nice, clear plan. We've got monetization, we've got this figured out. We understand what the market is saying. What are the competitors like, etc.? And so once, so once you've done this, guys, it only takes a minute to get through some of these things. Okay, you've gone through, you've asked for some feedback, you've asked for some more feedback from, from Grok. And so what you're going to do actually is you're going to copy the response from Grok and you're just going to drop it back into ChatGPT. Okay, it's going to be a huge long thing and just Shift+Enter, Enter so you can add more. And then you're going to even this second one, I'm going to copy it. And I'm going to drop it into ChatGPT. And I'm also going to delete a couple things that it said because I don't want it to do a couple things for me. And so, you know, obviously read it really quick. Just going to delete one thing there. Okay. And then you're going to before, before, you know, sending it off, you're going to go Shift+Enter, Enter again, and you're going to say, please create a— please, sorry, please revise— sorry, please revise the business plan document with these thoughts in mind. Whoops. Remember to keep my preferences and answers the discussions as the primary direction, but bring in the suggestions to improve on these ideas, make them more feasible. Okay, so I'm going to copy this prompt as well. Chat, paste, go. Okay. All right. So I'm going to go ahead and send this. And what that's going to do is it's going to rewrite this document. All right. So here it goes. You can see it start to do this stuff, but it's going to rewrite the document. So we've built a business plan. We took it to Grok. We asked for some feedback and some market sentiment and feedback over there. And then we bring that back into, into ChatGPT and we say, okay, hey, I want you to revise this. I want you to make it even better. While remembering what's going on.

Did you take both the deep search and the think outputs or just the think outputs?

I did. I copied each of them and I did Shift+Enter+Enter so that I could paste it all in one big thing.

Awesome.

Thanks, sorry. You're good. Please guys, please ask questions as you're going along. Get it right as you go. It's pretty fun building this stuff out. I love using AI as something to pull thoughts out of my head. I have ideas. I have so many, so many ideas, right? I'm right there with Drew. I've got 20 ideas that I can, I can build today. And it's really fun to have AI be able to kind of start to put them together and like make a plan and be like, hey, this is actually kind of good. Every time I build one of these plans, every week that I build it, I'm like, man, I want to do this. Like even crazy stuff. Like I was building one the other day for, you know, becoming a, you know, selling mortgage loans in, you know, like a better way than anything else that's out there on the market. And it was awesome. I think I could make tons of money if I ran ads to that website. Okay, so the business plan has been fully revised, Your Highness, incorporating all market feedback, strategic enhancements, etc. Okay, so now I'm going to look at the business plan, I'm going to read through it, and I'm going to kind of say, all right, first of all, I'm going to get rid of the timeline and the roadmap. I don't need, I don't need this stuff because my timeline is going to be 1 day, not 8 months. Scalability, I don't want to— anything it says related to, anything it says related to like development, you might want to take out of it. Um, I'm not going to do backend in Supabase. I'm going to build it in Replit. Backhand, TMDB. Yeah, so cool. TMDB API, Spotify API, so cool. Okay, so guys, how's everybody coming? Who has a finished business plan at this point? Anybody? I'm not hearing a lot of me's. Who would like 10 minutes to catch up?

I mean, I have something-ish.

Okay, how about we do this? We're going to take 5 minutes. We'll let you guys just go for 5 minutes to catch up here. So 11:05, and then we'll, we'll dive in. So go ahead and catch up, refine, read, whatever, you know, even those of you that are done, read your plan, refine it more, make it better, do whatever you can to keep improving it.

Hey, Jon. Yeah. This is asking me if it wants to— if I want chat to start building a prototype, would I prefer it as a React app or a simple static design?

Yeah, don't, don't have it build a prototype. Yeah, you don't have to build it. You just want to build a business plan and then we're, we're not gonna be building in ChatGPT.

Okay, perfect.

Yeah, we'll be doing all the building in Replit. ChatGPT, you can build code in ChatGPT, but not like that of what we're doing. Not even close. Hey guys, about another 2 minutes. Get to where you've got a business plan that has been— that received some feedback from another AI system, and then ask ChatGPT to refine your existing one. And that's where we'll pick up. Hey guys, let's go ahead and keep moving forward here. All right. The next step, the next thing we're going to do here, guys, is we've— so, so really what we're learning how to do is we're learning how to plan. Right? You can do this with anything at work, at home, whatever. You've got a problem, you've got an idea, you've got something you got to figure out. Just start talking to AI and work through the problem and have it interview you. And what are you trying to get out of this? What's the problem you're trying to solve? You know, whatever. And it's very, very smart and quick at being able to come up with ways to be able to do this. And it's drawing on all of human knowledge of, you know, from the beginning of existence. Right? It has, it has the ability to to do so many things. I've seen so many people turning to ChatGPT as their therapist because it can do therapy at a cost that's a fraction of what therapists can do or what they charge. And it's every bit as good. You can ask it to act like this. There are certain models out there that can be very, very good. And you can even turn it into a voice, right? You can even turn it on and say, hey, I want you to talk to me. I want you to interview me, right? And you could, you could just start a conversation with it. So these AIs are very, very smart. They're very, very capable. What we've done here is we built a business plan and that's great. So now what we're going to do is we're just going to close out of that. We'll come back to it in a second. The next thing we're going to build is a features list, right? So please build— by the way, I say please to AIs because, you know, someday they're going to remember and they'll remember. They'll remember that I was nice. Please build a—

They'll remember that, but they won't remember your Social Security number.

Exactly.

They're going to choose the things. The reality is this, Kelly, if they want my Social Security number, they can get it. That's the reality. It's already out there. They can find it. Please build a features list for everything we will need, both front end and back end for this app. When you say app, guys, it doesn't mean like an app on your phone. App, website, mobile app, whatever. That's just what it knows. I am going to say website though. I think most of us are doing that. We're just going to do that and I'm going to say, let's see, hold on. Please build this in a new canvas called Features Lists. That's what we're going to do. It's going to build a new one for us. Here's your prompt. All features list, right? So now it's just going to quickly build that out and it has this awesome, you know, here's the UI navigation and onboarding personalization and the home feed and the scene page and all these different things that have to be built. Okay, pretty cool. So go ahead and build that out, guys. Should only take a second for this one. This is a very fast one. And then it's time to get into Replit. So if you have not signed up for Replit yet, guys, um, this is the time to do it. Um, go ahead and use— I'll drop my, my invite code in here. Let's see here, where is it? Chat. Okay, that's my invite code for, um, for Replit. You get $10 off if you use that, and I get $10 if you use it. So I don't really care if you do, but Um, and it's just credits. It's not actually— they don't give me $10, they just give me credits for using the, using the app itself. But, um, yeah, go ahead and do that, and you should have by this point a beautiful features list and a business plan. Okay, all right. Time for Replit. The last thing that we're going to do here, guys, first of all, let me, let me come back over here to Replit. I'm going to— so we— I built this, you know, Cine Discuss, Cinema Chat, whatever. I'm going to leave this. I'm never going to come back to it. It's, it's good, you know, it's all right. Like, I could start to improve it and build it out, but what I'm going to do instead is I'm going to start a new one and we're going to do it right this time. Okay, so you just click Create App or go to Home or whatever and you get this thing. Describe an app or site you want to create. Now your description is going to be like this, can be done like this. You're going to say, give me a 3-sentence description about what we are building.

Okay.

Cinevia is a cinematic platform built for people who don't just watch movies, they feel them. Awesome. Okay. We're going to copy this. And we're going to paste it right here. And don't click Start Building yet, we're not done. Okay, so you're going to make your 3-sentence description and you're going to drop it here into Replit. That's your 3-sentence description. Then you're going to come back and you're going to snag your features list by opening it, copying it, coming back into Replit and pasting it. Okay, when you paste it it will show up as an attachment.

Excuse me, Jon, I was going to ask, should we modify it? I was going into the, the list and it says like, hey, utilize AWS and a bunch of different things.

Yeah, guys, that's a good, that's a good point. Um, if you do open your features list and it has a bunch of like extra things where it's trying to tell you exactly how to build all of this stuff, um, you know, delete it if you know what you're doing. If you don't, it's okay. We're gonna give it some instructions to build it inside of Replit anyway. Supabase, Firebase.

Yeah. I wouldn't necessarily know what to delete.

Yeah, I know. I know you wouldn't. And that's okay. That's okay. I mean, the things that it's doing, it's mostly doing is like a suggestion anyway, or if it looks like it's like hard, like if it's suggesting anything specific for a database or backend, just, just delete it. We just need feature lists, like we need an auth system. I don't need you to tell me how I'm going to build it. I just need to know that there is an auth system, right? Right. So we have a way to log in or log out of our, of our tool or whatever it is.

Okay. Sorry, I wasn't looking at your screen. So, so you got the feature list and you're copying You're pasting that as well on that.

Yeah, you're going to paste it. So yeah, so right here on Replit, you've got your 3 sentences and then I've pasted the feature list right here. Right. So it just shows up as like a little box, right? That's kind of an attachment. And then we're going to do the exact same thing, guys, with the business plan. Okay. So open the business plan, copy it, go back over to Replit and then paste it and don't click start building yet. That's where you want to be. So your screen should look like this. Right. Cinevia is a cinematic platform built for people who don't just watch movies. And then you've got your business plan pasted and you've got your features list pasted.

Okay.

What was your prompt to give me a 3-sentence description?

What? Yeah, it's in the, it's in the chat. Uh, no, it's not. Yeah. Just say, give me a 3-sentence description of what we're building. Okay. Yeah, I'll copy it in. I'll copy it in. Sure, thank you. Chat paste. There it is. Sorry, my bad.

And Jon, sorry, man, I keep interrupting.

No, keep going. Are we asking it to build? Yeah. As soon as you have your screen looking like this where you've got a business plan and a features list and then you've got your 3% subscription, go ahead and click start building. I'll go ahead and click it now. It's going to take a second and then it's going to come up with a secondary list of things that it wants you to ask, you know, answer about. Those additional features you can click on and say, oh yeah, that sounds cool, or that's cool, or, you know, just continue without them and you can add them later. But it is, yeah, it's going to create a comprehensive plan based on the materials you provided. So it's looking at, you know, both documents that you've attached as well as your 3-sentence description. It kind of knows what's going on. And we're going to use these documents more later here in the, here in the app. But for now, this is how we get started. Now it's come up for me, it's come up and said, here's the text app, app type, modern web app. That's great. Would you like any of these additional features? I do want, guys, if you do get to this stage, I do want a PostgreSQL database integration. Yes, that's what we're going to do there. Then we'll do some of these. Enhanced social features, we'll do that later. Apple Music API integration, you could do that.

NLP, Jon, what does that mean?

Advanced NLP? NLP is natural language processing. Got it. Yeah. So you could talk to it with normal words and it will understand what you do. That's what AI does. AI uses words and then understands it. So I'm going to go ahead and approve my plan and start. So this is where, this is where we want to get to, guys, on, on day 1. This is like the, the bare minimum of what we're able to get to. We still have some more things to go over today, but why don't you go ahead and tell me once you've clicked start build, start building, so that I can— you're done, Kelly, with the wine?

Yes.

Hey, yeah, mine started. OK, perfect. Started. Awesome. Started. Great.

Yeah, it's going very good.

I think I didn't hear David yet, or Kelly with an I.

Mine is.

Oh, yours is OK. Yeah, sweet. All right, it's going as well. Excellent. It sounds like we're all kind of rolling and this is really fun. This is where it's going to start to be fun. Some of you might have started to see a visual preview. It might have started to come up. Mine is, is not quite there yet, but it will be soon. So while this builds, you feel free to watch, you know, watch the thing come to life. And it's pretty fun to see this part. This is where it gets exciting. We're going to go over, we're going to do a 10-minute coding boot camp, maybe not even 10 minutes, might be 7 minutes. Okay. And this is the purpose here is not to teach you how to code. Remember, I don't know how to code, but having been a CEO of a tech company for, you know, a decade and before that another 6 years as, you know, the VP of— I was basically acting as the CEO of the company before that for about 3 years prior to that. And then before that, I was a VP and I wasn't acting as the CEO. I was more in the sales side of things. But bottom line is I've been around tech for a long time and worked with some amazing developers. And, and so I've, I've been in hundreds of meetings sitting at a whiteboard trying to figure out how do we put this together? How do we make this work? Oh, that's pretty cool. It used kind of like a Harry Potter feel for my site. I kind of love that. You guys are probably looking at your sites, but I'm looking at mine and it's cool. It's already way better than the last one. We've given it direction now, guys, right? This is so much better than just like build me a movie site, right? No, we've given it direction. Let's build a full business plan. So now when it's doing its chain of thought reasoning and thinking, when it's thinking about building a feature, does that, does that help the business plan? Yeah, it does. Okay, let's build it this way. Oh, that's, you know, he wants to accomplish this. We should probably do this and this and this, right? Like I'm already pumped about this, this thing here, right? Create my reel. I don't know what it's trying to say with create my reel, but I'll have to learn what it's built for me. But this is very, very cool. I'm excited to really play with this. Okay, but what we're going to do now is we're going to learn how to talk about code. It is very helpful to understand even something as simple as what's the frontend versus the backend. Understanding how an app or website works is very helpful. Again, you don't have to learn how to code, you just have to learn how to talk about it. Frontend, we already covered right at the very beginning. Frontend is what you see. It's things where you're— where you're interacting. For example, a login form where users enter their email and password. That's part of the front end, OK? And the front end interacts with the back end, OK? The back end is going to receive some sort of a call or an API or request or something and say, hey, you know, I just sent this person, just put their email and their password in. I'm going to check the back end. I'm going to check a table or a database in the back end and see if this combination exists. If it's correct, Then I'm going to send a signal back and let them into the, let them into the app, let them into the platform, let them into Netflix, let them into whatever it is you're logging into. OK, so checking the user's login credentials against database, return authentication token. That's one tiny example of what backend does. Backend really does all of the processing data, handles requests, it holds the data, communicates with databases. There's all kinds of APIs and calls and things happening in the backend and endpoints you can point at to work with other things. All you need to know is that the front end you can see basically, and the back end you can't, at least not to the average person. You will be able to as the developer, as the person in charge of everything, you'll have access to get into the database and look at everything and how things are happening. We'll talk about how that works. Database, guys, is no different than an Excel document. For example, will get a users table. I'm actually going to call this users table. This is a sample users table. You have the ID, the username, password, first name, last name, score, login, whatever, ID 1, 2, 3, 4, 5. We'll just create a new record every time you set up a new password or a user and it'll save all this stuff in a database. All that information will be there and you will be able to look at your databases inside of Replit and see how they work and understand how they work. But basically it organizes information, stores information. So for example, if you create a blog as part of your website, okay, you could— you can create— well, you can create it two ways. One, you could hard code it in the front end, so it's part of the— it's part of the actual, um, you know, website itself. That's kind of a really weird way to do it, part of the front end code. The better way to do it is to make a blog table, right? And that blog will say, okay, here's the new blog, here's this, here's the entry, here's all the, all this, whatever, right? Here's the formatting and here's when it was created, here's the author. And then the blog will dynamically load that content. And so if you change the database or if you add new content, more entries will be added into the database and more things show up in the blog. That is the proper way to do things. Um, most of the time. Most of the time, that's how you want to do that. Okay. SQL, you don't really need to understand. Know that we will be building a PostgreSQL database inside of Replit, right? And that's one of the reasons why Replit is really, really good is because it has everything that you need inside of one spot. Okay. All right. We'll go over tables and primary key relationships. We don't need to go over all that right now, but just know that those are there and We'll, we'll figure that out. You don't even have to connect database to Replit. Super easy. APIs and endpoints. What is an API? So an API is, it's a connector. It connects one thing to another thing. That's really all it is.

Okay.

It's like all the crazy cables I have here in my, in my studio for my music. Okay. I can just connect one thing to another and I have so many different cables to connect different things to another, except this is connecting software. So maybe, for example, you're building— like I was building a, um, uh, a website that would alert me automatically when my favorite rivers that I like to kayak come in, when the, when the levels get high enough, right? Which happens right now during the snow runoff, and I'm super sad that I have this cast on. I can't kayak anymore this year. I got all my kayaking in right before my surgery. But anyway, um, instead of me going out and setting up gauges— like, I could go out and I could set up a gauge you know, next to the river or whatever, and then set up a little Wi-Fi thing and, you know, a signal or a cellular thing and send it up, and I could have it send me a notification, hey, you know, the river's hit this level. Well, yeah, I can do that, and that's certainly technically feasible, but guess what? The U.S. Geological Survey has already done that for me, right? And they have tons of data that's available to the public, and so they have an API that I can plug into so I can see what the USGS you know, what level is happening on what river everywhere, right, all around the entire United States. And so I can use— I can just point at their data and then use it, right? And I can have it, I can have it, you know, request that data anytime I want, once a minute, once an hour, once a day. And, you know, when— as long as I— every time I do that, it does cost a little bit of money potentially. Um, the USGS one is free, the API is free, but sometimes some APIs cost cost a little bit of money. Usually the cost for APIs is a fraction of a fraction of a penny per call. It's very, very, very low. And so you can build very highly scalable apps that, you know, maybe you charge $10 for and cost you, you know, $0.35 a user for them and all their usage and all the calls and all the things they're doing. So APIs are not that expensive. A couple of them are, but for most of what you guys are going to be doing, they're all pretty basic. Okay, probably the most expensive API any of you guys will use will be Twilio, which handles text messaging. Okay? And it's still a fraction of a penny per text, right? But text can add up quickly. So you might end up being, you know, $1.70 per user per month or something that it might cost you for that if they're a heavy user. So APIs, endpoints, basically what we need to know is that from an API, you can do basically the entire internet. Everything that you click on, everything that you do does one of these 4 things. It either gets data, like grabbing that data from the River API, right? It posts data, sends new data. Maybe you're signing up for a new service, and so you put in your new username and create a password. It then posts that to the database. Where else have you heard the word post? How about on Instagram or Facebook, right? When you post to social media, you're actually posting. It's a technical term. You're writing to the database. That term came from developers. You're posting that to the database. That's what's happening, or you're putting it. Maybe you forgot your password and so you say forgot password, so it sends you a PUT link where you can update that data. It verifies you another way and then sends you a PUT link and allows you to update your password or delete. GET, POST, PUT, DELETE. This is often referred to as create, read, update, and delete or CRUD. Okay, CRUD is if you say like, I need a, I need a blog, um, blog section on the website with full CRUD functionality. Boom. That's way faster than saying I need to be able to create articles and I want to be able to update them and edit them and I want people to be able to see them and if I don't like it I want to be able to delete it. You can certainly tell the AI and it will understand you and it will say, oh, you want me to create CRUD functionality? It'll interpret that. But understanding these little terms, guys, will just help you understand what it's talking about and how to communicate a little bit faster. Authentication means we're going to let you in. Username, password, you can do a token-based authentication, JWT, which is what most people do. You can connect to systems like login with Google, login with Facebook, login with whatever. Those are set up through APIs. Replit will help you set those things up. It's very easy after you've done it once. You'll be able to do it again. We're going to host and deploy everything inside of Replit. We will then be able to point at your own custom domain. If you've bought one on GoDaddy or somewhere else, or if you want to point it somewhere else, that's fantastic. We'll worry about that. You won't have to use any of these external hosting services, although I will say that if you build it in Replit, you can certainly export this into GitHub and go do whatever the heck you want with it. It'll write 100,000 lines of code, 400,000 lines of code, whatever you're building, and you can then go put it wherever the heck you want. Okay, so that's really all you guys need to do. Um, you don't need to— I'm not going to have a quiz on this. This is not, you know, this is just for your benefit to say, okay, um, can we just maybe, you know, speak about this in a little bit more intelligent way? So right now we're still, um, we're still building storage, auth, routes, main, index. What I want to ask is, has anybody's app gotten to a point where it's asking a question, asking for an API, or it's finished. Is anybody there yet? Okay, everybody's still kind of in this building stage, and that's fine. It's going to build it out. It takes 10 to 15 minutes to build out this first step. In not too long— Kelly, did you have a question with a Y? No? Oh, sorry, you were unmuted and I just thought I heard something. Um, in not too long, guys, you're going to get to a point where it's going to say, hey, I want— we now want to connect an API to this. Okay, first of all, let me just ask this. I'll just ask at a high level, any questions about any of the things I went over in this coding thing? APIs, like, do you understand what an API is? Do you feel like you— what it means? If it's unclear, tell me. Or if, if you have a question about any of these things, APIs, guys, are going to be so— this is what makes apps so powerful is when you take something over here and something over here and something over here and you create this unique combination of services that, um, that no one else has ever done before, right? That no one's ever combined them in that way. And so you have this new, this new thing, this new idea because of all these ways you've connected it to make it, you know, that much more efficient for what you're trying to do. APIs is, uh, how the world runs in the software world.

Okay.

All right, guys, um, there's a couple things we can work on right now. Um, trying to think if I want to have this discussion right now or if I want to— if I want to wait a minute. Um, no, let's do it right now. Let's do it right now. Um, okay, guys, let's I'm gonna pull open the overview right here. Let's do, let's do this one. Um, nope, it's not this one. It is probably the advanced one.

Yeah.

Just trying to decide if I wanna go into this right now. I usually wanna go into it in about 20 minutes. We're going pretty fast, guys. We're doing pretty well. Even though, you know, you guys have felt maybe like you're holding people up or holding, you know, going a little slow, we've been pretty efficient going through this. I want to review what we've done so far. I think it's really, really helpful to just understand the types of tools, the type of process that we're going through. There's a lot of different ones you can do. Like we didn't, we didn't do a lot of other things. Like for example, if you're really building a business, it might help to get some real user validation, right? We only used AI to do that. Now, I will say that I think AI can do a pretty dang good job at figuring out whether or not something is going to be viable in the marketplace. And you have to ask it, right? You have to, like, you have to ask it 3 times. You have to say, do you think this idea will work? Yeah, I think it's good, right? Where do you think I can run into problems? Do you actually think this is a good idea? What do you think we can do to improve this idea to make it better for, you know, this audience that we're going after? Right? Why are you saying that? Do you have any research to back that up? Do you have any research that's been done? Can you point me to an article I can read? Can you point me to 10 articles I can read where you've done this? I really do think you can do pretty amazing research using the different AI tools that are out there, whether it's ChatGPT or Grok or Anthropic or, you know, Perplexity or whatever. Perplexity technically is just a conglomerate of or conglomeration of all of these different AIs put together. But, but yeah, the basic thing that we did, guys, here is we started with an idea, just a sentence, and maybe you just start talking to it about, here's what I want it to do, here's kind of the flow, here's the reasoning behind it. And then we talked to ChatGPT, we then bounced it out, we had it interview us, we then bounced it off of Grok. We brought it back in and created a few documents. Creating this alignment inside of Replit, you don't know it now, but as you start building out with this, and many of you will keep building all day today, I've had many people take this class and they're like, yeah, I didn't stop building until 3 AM. And then they show us this crazy thing they built because it just is so fun. Because what we've done now, we haven't even started really diving into, you know, starting to work on the user flow and the experience and everything that's happening. You know what I'm going to do here, guys? If you really like the way that your app looks right now, take a screenshot of it, because sometimes when it goes through and it starts building out more features and more elements, it's— it kind of rethinks everything and tries to outdo itself. And sometimes I say, hey, you know what, I really liked what you did before, I'm going to just drag this back in. Okay. This is building so many things right now. And yeah, it's really fun and awesome, but sometimes it'll get lost in itself. And I can't even, I can't even click outside of this thing because it's not interactive yet. All right. So here's what I want to do. We're going to go, we're going to play a video. We're just doing half minutes. Okay, so guys, pay attention here.

First, we believe as an industry—

pause. So this guy, his name is Eric Schmidt. He is the former— he was a Google CEO for about 15 years. Um, since he left in about 2020, he's been working for the last 5 years or so on a new big think tank thing, whatever. Big tech guy. Um, and he's pretty dang smart. So I want you guys to just listen to this, about 2 and a half minutes long, and then we'll talk about it. Whoops, I clicked play. I don't know why that did that. That play button looked weird though. All right, here we go.

We believe as an industry that in the next 1 year, the vast majority of programmers will be replaced by AI programmers. We also believe that within 1 year, you will have graduate-level mathematicians that are at the tippy top of graduate math programs. So that's 1 year.

Okay.

What happens in 2 years? Well, I've just told you about reasoning and I've told you about programming and I told you about math. Programming plus math are the basis of sort of our whole digital world. So the evidence and the claims from the research groups in OpenAI and Anthropic and so forth is that they're now somewhere around 10% or 20% of the code that they're developing in their research programs is being generated by the computer. That's called recursive self-improvement is the technical term. So, what happens when this thing starts to scale? Well, a lot. One way to say this is that within 3 to 5 years, we'll have what is called general intelligence, AGI. Which can be defined as a system that is as smart as the smartest mathematician, physicist, artist, writer, thinker, politician. I call this, by the way, the San Francisco consensus because everyone who believes this is in San Francisco.

It may be the water.

What happens when every single one of us has the equivalent of the smartest human on every problem in our pocket? But the reason I want to make the point here is that in the next year or two, this foundation is being locked in and it's not— we're not going to stop it. It's much more interesting after that because remember, the computers are now doing self-improvement. They're learning how to plan and they don't have to listen to us anymore. We call that superintelligence or ASI, artificial superintelligence. And this is the theory that there will be computers that are smarter than the sum of humans. The San Francisco consensus is this occurs within 6 years, just based on scaling. This path is not understood in our society. There's no language for what happens with the arrival of this. That's why it's underhyped. People do not understand what happens when when you have intelligence at this level, which is largely free.

All right, guys, thoughts?

When was that given?

Few months ago.

Uh, Sharon and I talk about this one a lot, about about being scared of any sort of technology. But I mean, technology could be a toothpick, right? And we aren't scared of toothpicks. Why? Because they can't outsmart us, or we feel superior to the toothpick. But it's still technology we had to integrate into our systems that change tooth health or whatever. So AI is another tool that if we pick it up and learn it the same as any other technology, any other surgical technique, anything else that we're trying to keep up with constantly in this world, then we can utilize it better instead of fearing the new technique, curing cancer sooner, essentially.

Toothpicks can't stand up and walk over and poke you in the eye though.

They can't. So that is true. It's scarier, but everything was scary at this phone would have been absolutely impossibly scary to someone 100 years ago, right? So it's the same thing. But if we aren't gonna learn it, or learn how to coexist with it, then you're right, it could essentially want to poke us in the eye. But there's no reason for it to poke us in the eye if it understands our value as well, which is that we want to use it as a tool, and it needs maintenance or whatever reason the robots decide to keep us around. Who knows?

Okay, perfect. Other thoughts?

Yeah, it feels like we're on the verge of another, like, renaissance per se. Like, it does feel bigger than, you know, like there's been some big digital milestones. There's like the internet, then there was like the mobile phone, then there's like Web 2.0. And it feels like as big as probably the internet itself just because of the general application of it. And so in general, it feels like if this is the— if we're on the brink of another renaissance, then like those who are able to like use that to their advantage are going to benefit at an outsized— to an outsized degree more than those who don't. So, you know, it's like almost like learning to read. And the more books you have at your disposal and the more you're learning to read, the more you benefit from the fact that now books are more widespread and you can gain knowledge and you can use that knowledge to do all sorts of awesome things while the general population is like still, you know, not able to read. So I think the question is like, how ubiquitous does it become? Because if this is something that's accessible to everyone who has the internet, and most people have access to the internet now, and you can cue it with a voice prompt, then really those who benefit are just those who are creative enough to apply principles and understanding to a voice prompt to then create the thing they want to create. Essentially, everyone will be able to create everything they want to. It's just that those who create the most usable, interesting, unique things are those who will gain the most, potentially.

Yeah, you bring up a really good point there at the end, Travis. We— what happens when anybody, you know, my 10-year-old son can create, you know, Calendly, right? Anybody can create any app that's out there. So why are you going to pay $12 a month for it or $99 a month for it? Why is someone going to pay Salesforce, you know, $800,000 a year for a team of 40 sales reps to have access to that when they can rebuild that in, you know, 2 weeks, get a little team together? Hey, what features do we need? And hey, we have the prototype working by the end of the first meeting. Now let's add some more features. Hey, you go work on that. You go work on that. You go work on that. Let's put it all together. A week later, you've, you've rebuilt Salesforce. It's gone, right? What does that, what does that do to so many businesses? The consensus is that by the year 2030, 400 million jobs will be lost globally because of AI. Okay, now there are some jobs that are going to be more proof, or, you know, AI-proof. Kelly, you're in a pretty good spot. Kelly with a Y. Mike, Andrea, Mike, you're going to have— AI is not taking your job for a minute. It could be decades before robots can do the types of things that you can do and going around to people's houses before people even socially will trust a robot to walk into their house and do therapy on them. You got to get over some mental barriers as a society before people can get there. So yeah, guys, superintelligence, self-recursive self-improvement. It's getting better and stronger faster than we can even begin to stop it, right? There's a great movie on this called Age of Ultron and what happens when it, when it gets out of hand. And there's other movies too, Terminator, etc. But what are your thoughts, guys? What does this— how does this change? How does this change your business, your industry, your personal life? With, with this type of massive societal change that's happening or that's about to happen? I think it— I feel like it's just starting.

I mean, for me, it makes it like even like AI charting, there are things— I mean, companies are slow to adopt them because they don't really add value to them. They just add value to the clinicians, honestly. And it's not really a cost saving besides maybe like burnout.

Out.

But, um, I've seen AI charting, and you can chat about just random things and it'll edit that out, it'll autofill the notes. And, uh, I see it making like the aspects of my job that I hate, which is all like the paperwork and managing my schedule, like disappearing for me.

Yeah, that's an interesting thought. Um, You can fast forward 5, 10, 15, 20 years, whatever it takes, even let's say it's 50 years, I don't know. But I don't think it's that long, by the way, to where AI can basically do everything at such an automated level when you add robotics to the equation that there's no legitimate reason why every single person on the planet shouldn't have all of their basic needs covered. Housing, food, healthcare, mental healthcare, everything that you need is available to you. A lot of people call this UBI, but it's UBI on steroids. When you can automate everything and hoping that you have good leaders in place to help make it happen, theoretically, you could get to a point where humans, the word Renaissance was brought up by Travis, Right? Humans now have to figure out, like, what exactly are we supposed to do now, right? What, what is the purpose? If all, if all of my basic needs are taken care of, what do you do when you don't have to work in the same way that you used to? What do you do with your time? What if you have a business? Maybe some of you have gotten to this point, right, and at certain points in your life or whatever, where you— all your financial needs are met, and so you're just like, what should I do? Like, what is my purpose in life? What am I supposed to do here?, right? Like for me, I'm asking those questions like, should I, should I go start a kayak school? Should I become a full-time composer? Should I, you know, what should I do? What are the things that I would really love to do? You know, assuming there wasn't anything there. Luckily I love AI and so I'm, I actually have a blast teaching these courses and talking to you guys and working with you guys. But, but yeah, it's an interesting, interesting discussion. Anybody else wants to, wanna say anything on this? We're gonna start, I'll say this, we're gonna start tomorrow. I'm gonna ask this same question again, right? Now that you've thought about it for a day, what do you think? How How does this affect your life? Okay. And I'm actually going to stop anybody else from talking right now because I want to jump back into Replit. I think we're probably pretty close here to being able to do some cool stuff here. Oh, great. Now mine's run into a technical problem. This happens sometimes, guys, where it will say, okay, I'm trying to, you know, get this to work. Yeah, fix it. This is what I'm going to say to fix it. Fix it. Okay. It's created all these wonderful things and then it did something at the end where it messed it up. And now I just needed to fix it. Just fix it, get it running, get the app running, make it work. But what I'm going to do instead, I know it's going to be working on this for a second. I wish I could switch to another computer and show another app right here. Hey, it looks like it fixed it really fast. This is fantastic. What you're going to do basically from here on out in relation to the app, we're going to add more pictures, we're going to add videos, we're going to add APIs, we're going to add all kinds of backend system things that don't exist. One of the things that we do need to do, and I want everyone to do this right now with their app, especially, well, unless you have already figured out that you don't need a database, but I think just about everyone's going to need a database. Right here, right next to the preview, there's a little plus button and there's all kinds of cool things that you can add and integrate here and so many cool things that we'll talk about later. Right now, we're just going to click on Database. Okay. And then we're going to click Create a Database. That's all you gotta do. Okay. Won't cost you anything. You just turn it on and the database is now active inside of Replit. Then you're going to say to Replit, um, I just turned on the database here in Replit. Please build out Or please connect the app to it. That's what we'll do. That will do it. Okay, I'm gonna go chat and I'm gonna go paste. I dropped that in the chat here for you guys. We want to get the database running. If you don't get the database running first, guys, it will try to build everything in memory or in what's called mem storage. Okay, which means if you try to create a username and password and things like that, or you need to log in or authentication system, whatever, as soon as you click refresh, it erases all the data that you do. If you create a database properly, then it will store all the data and you can have what's called persistence between sessions, right? It will retain all of your information, all the things you've done, all these fun things. Okay, don't send a message— usually don't send a message while it's in the middle of doing something. You want it to you want it to figure out those pieces here. It's fun also to read along with what it's doing. I have become a pretty decent developer that doesn't know how to code just by sitting here and reading for hundreds and hundreds and hundreds of hours of Replit working. I'll just sit here and I'll be working on something and be like, okay, what did you do? It helps me understand what's going on inside of my app. Even though I don't understand it completely, I understand, okay, you're working on something related to the client file here. There's some API request. I'm not sure why it's querying this. I don't know what this is, but, but now I've kind of seen it, right? Hey, this is trying to fetch something from here and it's having problems. All these endpoints aren't here, so we're going to add them, right? So it just kind of— it's going to work through a few bugs in the beginning to just kind of get it to a semi-stable first draft. And then what you're going to do, guys, is you're going to act as your own customer, right? That's all it takes. Now you are the user. You're a QA specialist now, okay? You're going to go in and you're going to click every button. And if you click on a button and it doesn't work, you're going to say, hey, this doesn't work, this button doesn't work, right? Let's build this page. Looks like you haven't built that out yet. Let's build— hey, I can't log in yet, it's not working. Oh, that's— let me set that up for you, right? And, and the reality is it probably just hasn't built that thing yet. It's not that it doesn't work, it just it just hasn't been built yet. Now everything seems to be working and that's awesome. Took a screenshot, it looks great. It's about to finish for me on this next step here. Here we go. Now I'm going to tell it I built the database, or I turned on the database, please connect the app to it. By the way, does anybody have any cool apps they want to show off? Your first draft? Anything look really neat? I'd love to, I'd love to see it if you have something that's cool.

I had something great, but then it ran into an error and I can't get back, get back to it. But it was, it's cool to see how it does the, I mean, really great front-end development. Honestly, it looks really modern. And I mean, yours looks great, obviously, but pretty impressive as someone who's paid a lot of money to developers to, you know, man, this stuff.

Now, obviously, we have to get it working, but that's Yeah, you, you just have to go through a few of this, like, hey, it's not working, hey, get it running, and hey, you know, whatever. It's going to run into a few just kind of problems in the beginning here, but it's literally just, you know, get it running. Yep, fix that, that's good. Let's turn on the database. So you kind of have to do a little bit of initialization here, but after, I mean, maybe after this step, I should be pretty close to being able to start working on things. Now it's not perfect, right? You look at my menu here, you know, some weird formatting, and there's a couple things that don't make sense yet. And, you know, hey, Hans Zimmer, Interstellar, that's kind of cool. That's doing pretty good here. Um, Cinevia Plus, $5.99 a month. If I click on that, it does nothing. So I just start to kind of learn, like, what I've started to build. Do I like it? Do I not like it? Do I want to work on other things? Um, what is this? I don't recognize that. I've got, I've got Twitter, which obviously doesn't exist anymore. Instagram. What's this little, like, basketball thing and this little cat thing? Does anybody know those? Anybody recognize those?

The cat one is GitHub.

That's GitHub. That's right. Totally is. I don't know why they would put that there. I don't know what the little basketball or whatever. That basketball, I think, is that little app I developed, went really viral. Shooting from— Got it. Perfect. Well, I'm glad that Replit figured that out and connected it. It knew you were on this course. Yeah, guys, it's fixing a couple of things, LSP issues, just syntax and things that it just didn't get quite right. Now it's starting to change things quickly and it's written a huge amount of code already. It's probably written close to 20,000 lines of code just to get this thing set up. That's just your base level. Then you're going to go and you're just going to start working on it. But the way that you do this, the way that you start fixing things is there's a couple of things you can do. One that I love is a screenshot technique where I'll do something like this. I'll say, and by the way, guys, on Mac, you go Shift Command 4, and that brings up this little— it's hard to see, it's a little plus, a plus thing. Can you guys see that? I don't know if you can see it very well. But anyway, I'm just going to screen— I'm just going to grab the nav menu here, right? I got it just in time right before the app said, you know, did a restart. And I'm going to say the nav menu looks kind of squished on desktop view. Let's make it a bit better.

Hey, Jon, quick question. That's what my app says too. It just says we couldn't reach this app. I mean, it took a screenshot and everything. It looked pretty good, but I guess what do I do?

Yeah, just say get the app running. Okay. Yeah, just say get the app running. If it says it is running, you say, no, I can't see it. It's not working. Oh, okay, sorry, let me look at it. Then what it will do is it can check and you can look right here, there's a little DevTools button right here. If there's a red number next to it, I can't click on it until it comes back up. But if there's a little red button there, it'll say that's your console and it'll say all the errors and things that are there. It can read the console, you don't have to tell it or copy and paste them in. You just say, check the console, there's some errors.. Right? So here I can click on it right here. Something's going on with this. And I, you know, I don't know. I don't know what this is. Warning. Something weird is happening. I have no idea. I have no idea what this is. I, I don't know what it means. Right? And so I can just say, hey, something weird is happening here. But if you just say get the app running, it'll, it'll figure it out. Okay? So now it's working on the navigation menu. So I did a screenshot of this. Okay? Hey, look, it's starting to move things around. Now it accidentally pushed the, you know, this button off the edge. That doesn't work. Another thing you can do here, guys, is there's a little button, the screen size button right here. You click that and you can look at it on like a phone to be like, okay, what's this going to look like over here? Okay, well, where's the menu? It's gone, right? There should be a hamburger menu or something like that, right? Login, sign out, whatever. It's working on the menu right now. And let's get rid of DevTools. There we go. But you want to be able to look at it at different, on different things. It should build a pretty responsive site already. Like honestly, Guys, this is pretty awesome. It already looks pretty dang awesome considering how long I've worked on it, which is a grand total of like, I don't know, 4 minutes of me typing. We've done some waiting, but I haven't done very much work. Now it's saying, hey, the nav menu should look better now. I'm going to look at full size. It looks like it's missing. I think it sucked at what I just asked it to do, but I am going to tell it the mobile, I can't see the Here's another thing that I'll do. I'm going to click Select. So this is right down here. You've got this Attach button where you can, you know, you can obviously drag over some screenshots, but you can also use the element selector. Okay. The element selector allows you to click it and then you can grab any element in the entire thing. So perfect. For example, maybe you don't like this, this text right here and you want to change it. You can't just click on it and change it. Okay. This thing, what you're seeing right now is a render of the code. So you can't just go on and change it. It's not a visual editor like that. You're going to click on it and you're going to say, change this tag or make this text punchier, right? And you can talk to it just like you're talking to ChatGPT. It's using the model underneath it, guys. So that you know that this one is using is made by Anthropic. And Anthropic is, you know, it's trying to bring humanity into AI a little bit more. And it's supposed to be the best, the best AI out there. That's what they're trying to be. And they are big time. These guys have raised billions and billions and billions, and they're one of the lead groups. And their lead model right now is Claude 3.7 Sonnet, and that is what is powering Replit. It is using one of the best models available in the world to do all of this stuff. And so it can do that. So I just had it change it. Hey, unlock soul-stirring soundtracks and join a community that resonates. So it made it punchier. Okay, great.

Awesome.

Now I'm going to have it go back or I'm going to say I'm gonna click explore scenes. Hey, this button, this button doesn't work yet. Let's build it out. Okay, awesome. I'll do that. So now it's gonna build new pages so I can go click on the scenes and explore them and figure this out. And how are we gonna do this? Right? So this is, this is what you're gonna be working on now, guys. You're acting as your own user. And you have to go through and you have to click on things and say, okay, this, let's build this system out. Let's build this system out. Let's make this work, right? Until you can click through the entire thing. So I built, guys, I built all of this on my, my own website here where you guys signed up for the course, right? GenAI API. This entire thing was built in Replit by me. I have not had one other developer, nothing. Okay? I've had nobody come in. I have done all of it. Okay? I built the authentication and the admin systems. And the, you know, the live courses and the course content and, you know, all of these different, you know, quizzes and all of these different things. Every single button works. When I go down to, you know, privacy policy, right, I built out, you know, this entire thing with AI, of course, right? Built a beautiful privacy policy. Um, this email address, privacy@genaipi.org— I'll show you guys emails tomorrow. We won't get into that today, how to connect email systems up and how to set up business, that part of the business, right? And how to only pay for one email address. I don't pay for privacy@genaipi.org. I only have jon@genaipi.org. Um, by the way, guys, if you could do me a favor, if you— I, I, it— I think that the jon@genaipi.org emails went to your junk email. That's my guess. If you guys wouldn't mind, please click on my email and put it in your inbox. Mark not as spam, mark as safe, whatever it is in your system. I would really appreciate Little side note, help me get my deliverability better. This is a new domain, and anytime you do a new domain, it's only— it's a couple months old and I've used it quite a bit. But anytime you send out a message to like 8 or 9 people from a new domain, it doesn't like that, right? It thinks that's spam, right? And so anyway, so what your, your job now, guys, when you're building a new business in 3 days, and a lot of what, what people have reached out to me for to learn about this is to learn the AI skills. Yes, that's great. But also, hey, I want to, I want to start generating money by the end of the week. I want to generate money right now. Well, if you want to generate money with your business, what you have to do is not die a death of a thousand paper cuts. If you— if people come and they click on your about page and it looks weird or the link is broken, or they come in here and then they click on this button and it doesn't work, then, you know, you're, you're, there's a little paper cut, right? They might not, they might not leave the site. They might be like, oh, that doesn't work, right? And they'll come over here. But then if they click request a demo and it doesn't work, right? Or, you know, they click on one of these buttons down here, they click on the, on the Instagram and it goes to nothing. It doesn't exist, right? Whatever it is, right? Those are all little tiny paper cuts. And after you have 3 or 4 paper cuts on a site, you kind of just leave. You kind of don't feel like giving them your information anymore. You're like, this site kind of sucks. These people aren't really paying attention to what they're doing. But if people can get through your entire experience from, hey, I'm here, I can sign up, I can log in, I can buy, I can click, I can, you know, everything works, everything I click on seems to make sense, the forgot password button works, you know, I click on my profile, I can look at different things and all this stuff works. Yeah, see, this doesn't work right now. I did that on purpose because I just barely fixed this and I knew it didn't work and I haven't deployed it yet. But if I come across 404s, then I, I don't know, that's a paper cut for me. And I'm like, oh, hey, what the heck's going on? Like now all of a sudden my brain is like a little alarm bell goes off. If you want people to put their credit card into your solution, you gotta get rid of the paper cuts. Okay? So that's what you're gonna be doing inside of Replit is going through each little thing. Now if I click explore scenes, okay, now it does something. That's cool. The explore button is here. Now what I'm going to do is this, I am going to screenshot this. Screenshot and I'm going to say this happened when I click the button. Also, I see all these errors up here, so something's happening there. Also check the console for errors. Errors. I'm going to click that and I'm going to say, hey, that didn't work. Now I'll go back right now and I'll keep messing around and looking around and reading and I'm start to take notes while it's working on something. I am always like, you know, I'm gonna, I'm gonna click on, hey, does timeline work? Oh no, that doesn't work. Audio work? That doesn't work. Like, what, what is supposed to, what is supposed to happen here? And sometimes I don't even know what's supposed to happen here. And so I ask the agent, I'll be like, hey, this feature, I'll go select right here, you know, this, this section here. Yeah, what is this timeline button? What is this supposed to do? What is the purpose of it? Oh, this is what we were thinking here. And you're like, yeah, get rid of it. Or yeah, great, let's connect that. Let's build that. Right. And it will just build it out. But you're just going through and you're going to start building these features one bit at a time until, yeah, I got to figure out how to get scenes in here. Am I going to use YouTube links? Am I going to actually upload video and audio, which could be really heavy? I probably am going to do YouTube links because that's going to be using somebody else's service to host all of that stuff. All right. So Yeah, yeah, there's some profiles, meet the community, find friends. I guess I'm starting to kind of understand what's going on here. There's obviously a login, there's clubs, my reel. I still don't know what my reel is actually supposed to be. We've got an explore button now that's starting to think. Um, here, let's click this and unclick it. There we go. Hey, it already fixed it! It's still working, but now I can click on scenes. Look, it built this cool new scenes thing. So I'm right there with whoever said that a second ago. I think it was Drew, maybe. Drew's raised money or been in a business that's now sold. And I've raised a lot of money and I've paid tens of millions of dollars, guys, to developers throughout my career. And I've never been a coder. And so every time I come up with an idea or something I want to do, I have to go raise a bunch of money. I have to say, hey, Dad, can you lend me $20 grand? And here's, here's my plan to give it back. And, you know, guess how risky that $20,000 is that you borrow from your dad or your uncle or your neighbor or whoever you raise money from, right? I've raised money from my dad, but I've also raised it from 65 other people, right? I've raised it from a huge amount of people. And that puts a massive amount of pressure and stress that's unnecessary on you as a person and as an entrepreneur. And that's just kind of been part of the game. That's been how— that's been how it works. You got to get investors if you want to go build something that's big, or you got to save a ton of money and then risk it all to go get something made. And now with AI, with tools like Replit and other ones that are out there, but this process that we're using today, you can validate an idea, at least on the front end, build it out, get it completely working, and then go put it in front of customers, run Facebook ads, just go to the mall and hand it to people, knock on people's doors and show it to people, whatever it is, find your target audience and get it in front of them and try to get them to buy it. And you can test with live product right there without having to wait. It's incredible, right? For you, you run a $100 Facebook test, throw $10 a day, $15 a day on a Facebook test, put your target market in and just run the ad. See if anybody buys it, right? If anybody buys it, call them up. Hey, I'm the founder of so-and-so. Thanks so much for signing up. Why did you buy this? You know, talk to them, whatever. But you can do that now, guys, for a matter of $100. It used to be $100,000. Now it's $100, right? We have reduced the cost of that early stage riskiest part of entrepreneurship and even building new things inside of an existing company like, like Wally or like Boostly or whatever, right? You, you can pay, you know, some offshore dev team or bog down your super expensive dev team or whatever it is, right? And, and be like, hey, we want this new feature. No, now you can just whip it up. Hey, I built this, right? It's ready to go. Let's go test it with customers.

Amazing, right?

I, I hope this first day, guys, has, um, it got— went by pretty quick. And I hope, I hope this has started to open your mind a little bit. Tomorrow we're going to, um, We're gonna do some fun stuff, do some image generation. It's always fun. Tomorrow we're gonna do some APIs, APIs, image generation, and more advanced stuff inside of Replit where we're gonna really start to build out those experiences, debug, you know, get things working, build the systems, check the databases and make sure they're working in the way that we think they are, understand how to look at the database and how to really make that stuff work. I encourage you guys today, feel free to get into Replit and just keep going and keep talking and keep refining and all this this stuff. Um, but, but just keep, you know, do your, do your best to, um, to, you know, dive in and build your dream, right? Um, you, you're all smart enough to do this without me, um, in terms of, you know, that you're going to run into some things and you're gonna be like, Jon, help me, right? Um, but really, most of the time, you're going to come to me tomorrow and you're gonna say, hey Jon, I was trying to do this and I couldn't figure out how to do this. And I'm going to say, did you say those exact words to Replit? You'll be like, oh no, I didn't. Then you type it. Any question you would ask me, just ask Replit. It uses that, what we were talking about. Someone asked about, I think it was David asked about NLP earlier. But natural language processing, it understands you, it understands what you mean. Just talk to it. Have good conversations with it, go build. So thanks so much, guys. I'll send this recording later in case you want to check it out or, you know, go back and watch something again. But we'll see you guys all tomorrow morning, 9 AM. Thank you. Okay, thanks, John.

See you guys. Thanks, John.