Testing 4 New Ai Coding Tools: Bolt vs V0 vs Replit vs Websim.AI

A year from now, these apps are going to be like, absolutely mind blowing. Like, you probably will be able to prompt a whole SaaS company into existence. Right?

Yeah. And the kind of stuff that you'd have to be a developer before to do any of this, right?

Yeah, yeah.

So if you're in business, I would be thinking about what that means in terms of the opportunities. Like, what are you going to be able to do with this new technology once it gets that good?

The way I've been using these tools is to create just like really basic, simple mini apps that sort of clear bottlenecks my business.

Hey, welcome to the Next Wave podcast. I'm Matt Wolf. I'm here with Nathan Lands. And we're entering into a world right now where the gap between having an idea in your brain and actually seeing a working app, a working piece of software on your computer screen, that gap is closing quickly. And we're entering a really, really exciting time right now where anybody can write code to do pretty much anything they need it to do. And there's a whole new crop of up and coming AI coding tools that we've been hearing so much about, and we wanted to put them all through their motions. Tools like bolt, tools like V0, Web, SIM, Replit, AI. These are some of those AI coding tools that we're starting to see everywhere and people are just raving about them. So we wanted to find out what they're good at, what they're not so good at, and really put them through the motions and see what we can build with them. So, so let's go ahead and just jump right in and play with some of these up and coming AI coding tools. We recently did an episode with Riley Brown where we walked through building out an entire app step by step with Riley using mostly cursor for that project. But there's been like this new crop of AI tools popping up, like Bolt New and V0 from Vercel and what was the other one?

Replit Agent.

Yeah, right, yeah. So that we've got all of these new AI tools that are popping up that are just making it so much easier to code. And I mean, like, even easier than what we saw with cursor. Right. Like with cursor. Even when Riley Brown showed it to us, we kind of started with a template and then we had like a starting point and then we use cursor to build on top of that template that Riley provided us. Right. Well, with these other tools like bolt and V0, you don't even need that template. Right. You Just go in there and you say, I want an app that does X and it just builds it and it gives you a working app. So that's what we want to do on this episode, is play around with some of those tools, put them through their motions a little bit, talk about the implications and see what we can get them to build for us on this episode.

Yeah, I guess let's just jump into Bolt.

Yeah. So Bolt is, like, really simple, right? This is, this is the screen you get when you first log into Bolt. You can come down here and choose what you want to use for your code. Right. You've got Next JS and Remix and Svelte and React.

I use Astro. Most people only know Astro.

It's like, I don't know what Astro is.

Yeah, yeah, yeah. My website's all I coded in Astro, which is like. Astro is a relatively new framework, which for, like, static websites is probably the best, like, because it loads the least JavaScript and so it's the fastest to load this kind of stuff. But it's, it's, it's, it is a bit intimidating to learn. So it's awesome that they have something like this here.

Yeah, well, this one, you don't even need to pick which sort of, you know, framework you want to build on top of, but you can just, you know, give it a concept and it'll build it out. I like to ask it to build a game first because I always feel like games are a lot more complex than what, like, most businesses are going to use it for. And it's sort of like a really good demonstration of something more complicated than what you're probably even going to use it for. Last time I did it, I had it build Tetris. We see everybody on Earth use the Snake example. So I don't want to do Tetris or Snake. I was thinking maybe like Space Invaders.

Let me have it.

Build Space Invaders real quick so I can just give it the prompt, build a Space Invaders game. And then just like, that's it. That's the whole prompt. 1, 2, 3, 4, 5 words. And then you can see on the screen, it's just coding. I'm not touching anything. It's going and writing all the code. You can see over on the left, it's already, you know, got all the pages and everything's already built and, and ready to go. It's just, it's, it's just creating it. I'm not doing anything. And let's see, let's see if just one prompt is enough to make this work. We can see here over on the left it says, I'll create a modern, beautiful Space Invaders game with React and Typescript. It will feature smooth animations, particle effects, and responsive controls. So here it says, I've created a modern Space Invaders game with the following features. Smooth canvas, particle effects, score tracking, modern ui, responsive controls with keyboard input. Beautiful start and game over. Screen performance optimized with proper frame timing. So you can see over here on the right of my screen, we've got Space Invaders. I can click start game. I don't know what the controls are. Okay, so I got a square that moves.

So it didn't really work on the first try, which happens often with AI. So I'm just gonna give it the prompt. All I see is a square that I can move around on the screen. I'll just give it that little amount of info and let's improve the game. Now it's gonna go on and code some more.

Random side story, but this is reminding me of back in the day when I launched my startup, demafy.com, we had this, this waiting list. And when you would join the wait list after you. After you gave your email, you would be prompted to fill out this puzzle, which was like using our logo. And you had to put the puzzle pieces incorrectly and it timed how fast you could do it. And then we had had a leaderboard and everyone was like sharing this. It was like this, like crazy viral things. It'd be kind of cool to like replicate something like that, where you have like a little, little game and use that as a marketing tool of some kind.

We'll have to see if V0 can do that when we switch over to V0 to test that out. Right now it's just going through writing all the new code. You know, there's a whole bunch of different files that are already generated and it's just jumping around between the various files and fixing the code. You can already see here it's 1, 2, 3, 4, 5, 6, 7 different files. It's already worked on since I gave it that last prompt.

Yeah, it's so cool that it gives you all the. All the code and like, does all the folder structure and everything for you. Like, I remember some of the early AI website generator apps or whatever, or a lot of them didn't give you code. Like, you just see like the final product and it's like, okay, now what?

Yeah, yeah, yeah, for sure. So here it says left and right to move, space to shoot. So hopefully this time we have a Working game.

A little bit closer. Apparently I died already.

I think this may be complicated for you. It needs something.

Yeah. Maybe Space Invaders is a little bit too complicated. I know I can get there with, like, two more prompts, probably. But what I really wanted to talk about on this episode was, was the concept of, like, just creating little mini apps for your business. I think a lot of people see these tools like Cursor and bolt and V0 and replit and all of these kinds of things, and they think, well, I'm not trying to build a software company. I'm not going to go and launch a SaaS. That's. It's just, you know, I've already got my business. I don't want to go do that. Right. But the way I've been using these tools is to create just, like, really basic, simple mini apps that this sort of clear bottlenecks my business. Right. Like, one thing I was running into is a lot of times I'm pulling images off the Internet and they get pulled as. As webp files, right? Like, if you generate an image in dolly inside of OpenAI and you download it to your computer, it downloads it as a Webp file. I really hate Webp files. They're kind of annoying to work with. I want it to be converted to a JPEG file. I can't use webp files in DaVinci resolve when I'm editing my videos. To overlay them onto videos, I need JPEG files, right? So I created a simple little app with a single prompt where I can drag and drop as many Webp files into that. Into that little app, and it would just convert them all into JPEGs, like, instantly. In fact, here, let me. Let me actually show you the little app that it developed. It just created this little Python script called webp Converter Py. I can open it and it opened in the wrong screen, but you can see it here. It just says webp to JPEG converter. Here's where the files will be saved. Code slash, Webp converter slash, saved. That's the folder right there. I made it so you can, like, change the folder, pick wherever you want them to automatically be saved. And then I've got. Let's. You can see I got these three Webp files that I just dragged over here.

One's a picture of a tongue that I downloaded for some reason. Here's a picture that's not your tongue.

Okay, good.

I think this is like a dolly. Yeah, this is a Dolly image, and this one's also a Dolly image. But. But they're all webp files. And so if I grab all three of these, just drag them and drop them into this little app that I built here, and it says, conversion complete successful three failed zero. And then if I look at my saved folder here, it just saved all three of those as JPEGs here. And like, really, really simple app. I know there's other apps on the Internet that already can do that kind of stuff, but I wanted to be able to just do it in bulk really quickly without, you know, ads all over the screen and all that kind of crap that you get when you go to those, like, online file converters. And so I just, I had Bolt build me this little simple app really, really quickly, downloaded the files onto my computer. And now, like, if I'm working in DaVinci and I'm trying to find, you know, images to overlay into my videos, I just have this little thing open on my screen. Whenever I download an image, I throw it in there. It gives me the JPEG back. It just saves me a few extra seconds on, on converting those files. Because the way I used to do it was I would open it up in, like, the Windows Image Viewer app, and then I would re save it as a JPEG file. And this is just like, way quicker. So that's the way I've been seeing these. These apps is just like little ways to sort of solve bottlenecks in my business. Like, what's something that I just, I want to solve this little problem. I don't really need to give this app away or sell it. It's just for me. And I found that really, really helpful. So, know, I think anybody can look at their business and go, where are the bottlenecks? Where are the things that slow me down just a little bit? Can I create a little quick script that's going to make that thing just slightly easier, speed up the process just a little bit. And that's what I'm loving about these tools, is just like finding those little things that just make my life a little bit easier. Yeah.

And that kind of stuff that you would have to be a developer before to do any of this. Right. Or, you know, now if you're. You could be a CEO and have an idea of something that make your business a little bit better, a tool, a simple tool, than maybe before, you'd have to hire an entire team of engineers to build it. And now if it's something very simple, in theory, you could just use something like Bolt.

Look, if you're curious about custom GPTs or are pro that's looking to up your game. Listen up. I've actually built custom GPTs that help me do the research and planning for my YouTube videos. And building my own custom GPTs has truly given me a huge advantage. Well, I want to do the same for you. HubSpot has just dropped a full guide on how you can create your own custom GPT and they've taken the guesswork out of it. We've included templates and a step by step guide to design and implement custom models. So you can focus on the best part, actually building it. If you want it, you can get it at the link in the description below. Now back to the show. Let's try something else inside of Bolt though, because it is actually a lot more impressive than this here. I'll even. Let me show you real quick some of the other stuff that I already built first because I thought this was really impressive. So I built this Tetris game. Here you can see, here's the preview screen of it when it loads. And it's just, you know, it's Tetris like you'd expect it, right? I can move this around, I can drop it real quick and it looks really, really good, right? So, you know, I'm just obviously like just dropping everything really, really quick. But it works, trust me. Like the lines actually all the scoring works. So you can see that that worked. So that was one of the apps that I created. This other one was just like a little fun one that I was playing around with, which is an. An ASCII generator, right? So what you do is you drop any image in and then it recreates that image as like words and letters. So let me, let me find an image real quick here. In fact, I can use my little. Let's use this tug image just because I was showing that off earlier. I want to convert this tongue image into text. So I'll just drag and drop it into this little like art generator here. So. All right, here's an, here's an image that I made of Donald Trump for one of my YouTube videos. So I'll go ahead and upload that and then you can see it converted it into like a text version of it. So just kind of like some of the random stuff that you can do with it. But yeah, do you have an idea of like another app we can test real quick?

I think it'd be kind of cool to see what it does when you give it to a, like a screenshot of a SaaS app or something. Like, can you can it copy an app, like from a screenshot?

We could try. Do you have a specific website in mind?

How about Linear? Linear?

Yeah, like linear is that linear.com.app. so let me just take a screenshot of like what we see above the fold here.

Yeah, I think so. Let's just see what does it do? Does it have any idea what to do? Or is it like totally off or is it like close and the graphics are just wrong?

So I just took a screenshot told to recreate this page. So let's see what it does here. I'm not even familiar with Linear, so I don't know what we're. What. What it even is. All right, so now we've got a screen that has got some text on it. Let's see what should I. What prompt should I give it to? Try again. Let's see. There are no images. This looks nothing like the original image I gave you.

Please make it better.

Let's just see what it does or you're fired. Add that and see if it works or you're fired.

Yeah, I'll try that next time. I'll start threatening it with the next prompt. Prompt.

No, just went back to just a white. Oh, wait, no. Okay, now we've got some really crappy looking gradients in the background so far. Bolt seems to be pretty good at making games and like novel stuff.

So I just told it like, create me a simple directory website for lore.com. make it like an educational site about the lore of different fantasy worlds. So this is like. And then make it beautiful. Just simple kind of website. Here's what Bolt gave me. So it's like, kind of works.

And obviously if anybody is watching this or listening to this, and it seems like this worked really fast, we are skipping over a lot of the, the actual writing of the code in the video just so that it, you know, speeds up the flow of this video here. So, yeah, I think Bolt is not amazing when you try to upload an image and try to get it to like duplicate that image. That's what I'm finding out. But I think if we were to just like explain what website we want, Bolt would do a pretty decent job. So now, Nathan, you're sharing your screen. You didn't give it an image this time, right? You just kind of told it what you wanted.

Okay, yeah, I was just thinking like, you know, let's imagine I was using lore.com for something different. Like I was using it for what a lot of people imagine would be like a directory about Fantasy or video games or like the lore behind different fantasy worlds and stuff. So, right, I said create me a simple directory website for lore.com, make it an educational site about the lore of different fantasy worlds, etc. Use Astro, which is a framework I like to use for static websites and make it beautiful. So this is what Bolt gave me, which I don't think is very beautiful.

I mean, hey, it's a lot more beautiful than what I got when trying to start with an image.

And here's what V0 from Vercel gave me, which I think it's simple, kind of minimalistic, but I think it does look better.

What I think the big difference is is like Bolt didn't pull in any icons, it didn't pull in any images, it didn't like do any of that stuff. At least when you used v0, it actually added some icons to try to make it look good.

Yeah, yeah. And it also, it's not showing it here. I'm trying to get to reload. It's not doing it again. But when it first loaded, all of this was like animated too. And it was really fancy looking.

Oh, nice.

I thought that was cool. Like it like had some basic animation, like all these, like the words slowly came in, you know, and like it like flashed in all that kind of stuff.

So yeah, yeah, V0 definitely has like a lot better design to it. It's just, it's so fascinating to me because when I was using Bolt the first time, like everything I was doing was working on like the. Either the first or second prompt. Sometimes I had to get it a second prompt and then it started working really well, but it was always within two prompts. I got what I was looking for.

I also tried this on Replit and so their approach is like really different because I think it's interesting. It's almost like you're. It's almost like you're chatting with like an AI project manager or something. Right. Where instead of like, you know, Bolt and V0, just go ahead and kind of build it for you and don't really ask for additional input. Like this is asking me like, it's wanting to clarify what I want.

Right, right.

They, you know, they asked me other things. Would you like any of these additional features? We can also make changes later. So don't want to add interactive timeline features. Sure. Don't want to implement character relationship maps. If they could do that, I don't know, that'd be awesome. I want to see if they can actually do that, add a user contribution System. I'm just going. I'm going to say yes, all of this.

Can you do me a favor real, though, and just, like, read out the prompt that you gave it just for anybody who's listening on audio? I want to make sure that they kind of have a little more context. And I also. It's very small on my screen, so I can't read it.

Yeah, yeah, yeah. The same prompt I used for other one, create me a simple directory website for lore.com, make it an educational site about the lore of different fantasy worlds, et cetera. I told it what framework I wanted to use. I said, use Astro and make it beautiful. So pretty straightforward. And. Yeah. And then it gives me all these options. So let's. Let's approve the plan and start. So it seems like out of the three, that V0 gives you the most beautiful stuff out of the box. But I do really like how it works on replit. In terms of just the whole interface of it telling you everything. Like, maybe for some people this is probably overwhelming because you actually see the. You see the code flying in here. It shows you as it's doing it.

Yeah, I mean, Bolt did that too, though.

Bolt.

You were able to, like, basically watch it create every single file and then write the code. Right. This looks prettier.

Yeah, this looks prettier. And I do like the whole interface of, like, it's like a project manager, where it's like asking you the different features you want. Right? Where. Yeah, yeah, yeah. And I'm testing all of them before. Like, I didn't test bulb, I tested V0 and Replit. Like Repl. It's way more thorough. Like, it'll give you a whole freaking, like, login system and stuff.

Yeah, yeah.

I mean, this is doing. This is doing the kind of stuff that, you know, that Riley was showing us with, with cursor, in terms of, like, creating a database and everything. Like, this is doing all that. It just created a postgres database.

It'.

It's incredible. Now you can tell the website is very ugly. So here's what it gave us. Fantasy lore. So it did a lot of stuff. It created a database and everything.

Well, it did write a lot of. It did create a lot of files for you. You know, maybe since I was sort of giving Bolt two tries, two prompts, maybe maybe the next best thing is give it like one more prompt. Say, like, hey, here's what's going on, and see if it'll just fix it for you.

Yeah, so the. So the, you know, the AI project Manager said, is the website displaying. Displaying properly with a fantasy theme and navigation working?

I mean it says the words fantasy. Is that enough to consider it a fantasy theme?

Website loads but there isn't any navigation. You can't really do anything with it. I don't know if that's enough but I guess let's give it like two or three tries and see where it gets us. But this is the one I use with my son like where we create like a little simple website like a. Giving like tips, tips for Minecraft and League of Legends. And it, you know, after like five tries got us something that was kind of okay.

Yeah. The other one I was trying to think of was Windsurf Editor. I don't know if you've heard of that one yet. That one to me looks very similar to Cursor. It looks very. Because it's also, I believe it's also a fork of VS code.

Just like Cursor was so, so replit. I think the agent's totally confused now. It's like is the website displaying the three fantasy worlds, Middle Earth, Westeros and Hogwarts Universe with their descriptions and featured articles? We're still so like they've got some kind of database set up but like definitely nothing is being pulled in.

So I think from all of the various apps that I've tested so far between bolt and V0 and the replit one, I actually think probably the best one at actually making like a fairly decent looking user interface has been web sim. And I don't think most people actually see websim as like a code writing tool, but it totally is. So if I use that same prompt that you created, I could come up here. What would you like to create today? Let's go ahead and paste in the same one. Create me a simple directing website for Lore.com, make it an educational site about lore of different fantasy worlds, et cetera. I'm going to get rid of use Astro because I don't know what language this. I think it kind of uses its own thing but let's go ahead and just prompt that and see what Websim creates for us. I don't totally know how Websim works behind the scenes, but it is writing a whole bunch of code for us behind the scenes. We can see it's still loading. But you could see as I move my mouse around it created this like little star field of the background lore. Explore the vast realms of fantasy, Middle Earth, Westeros, the Witcher's World, Hogwarts. So it created all of these, like little.

Descriptions and let's see. Oh, this actually went to your real lore dot com. But let's see, let's. If I go and click on like.

Worlds, where does it take me? Oh, gave me an error.

So these all have a lot of work.

One thing that you can do with Websim too is like, not give it any context and just tell it. Go like lore. Com. Right? If I type this in, it's not going to go to your Lore. Com. It's just going to create a whole new website called lore.com and just sort of decide what it wants to be all about.

Maybe they're getting hung up on this one because, like they think they need to create all this content and that just like gets complicated for it, like figuring out all the different images or text or whatever.

So this is what it thinks lore.com should look like. Journey through the most captivating fantasy worlds ever created. Discover ancient mysteries, legendary creatures and epic tales that shape these magical realms. And then if I click on begin your adventure. Let's see where it takes us.

That's kind of fun, like having it imagine what it thinks it would be. I'm trying to add another ones too, like crate.lore.com. like, what is that?

It actually kind of created the same content. Middle Earth, the Witchers continent and Westeros. Well, you can see it created like a fairly simple design. But if I come up here to these three dots over here on web sim, you can actually see there's a View source button and a download button.

Right.

I can click on View source and it'll show me all the code for the website that I can just copy and paste it to wherever. Or I can click download and it'll just, I believe, download it as a. Yeah, just download it as an HTML file. So this will actually write the code for you and then let you have the code. After it writes it. I'm going to give it another prompt, make the links clickable if I should.

Show this or not. But like, so V0 basically looked@lore.com and copied parts of it, like including the styles. So you could do that with other websites. You definitely could copy their styles just by typing that in. Bolt didn't do that. Bolt like tried to think of what lora.com would be. And it's got like some kind of like course platform with like, you know, 1,000 plus active courses and it's got featured courses below and all this kind of stuff like ancient philosophy, modern literature, world history.

Yeah. Okay, so now if I go To Lore Worlds. You can see it brought up Middle Earth, Westeros, the continent and Wizarding World. And then it's got like tags underneath. Let's see, if I go into Middle Earth, it gives me an error. So it's, I mean it's, it gave us a ui. It's not giving us like a full working website, but I mean it's one of the better looking UIs that we've gotten out of some of these platforms. But let's go ahead and jump over to what you've been getting. So you said you've been working with Vercel?

Yeah. So I tried what you did like create lore.com and I just wanted to see what it would come up with if you just literally just put in a domain and that's it. So here when I told Vercel or V0 just to create Lore.com, it actually went and looked at my website and basically copied the style. I mean this is not exactly. But this is the same colors, the same font, the same menu, even the same text for the button. You know, they, they even got my logo and made something that's somewhat similar.

Which is kind of wild. So this, they're definitely, you know, you, you could definitely look at the style of a website you liked and say, hey, let's like, you know, create something like that and, and you could do that. So that's, that's exciting. A little bit scary that anyone could just probably, you know, pretty soon just copy your website like that. So bold. When I, you know, say create lore.com, it didn't look at the existing website. It definitely like just went off and decided to be creative and like, you know, decide what it thought that would be. It's like it's created a course website here. I guess it says I have over a thousand active courses and over 50,000 students, which is great.

Yeah, congratulations.

Yeah, maybe that's what I should be doing.

And it's got all these like featured courses down here, right. Which is actually got like images and stuff. So it's, it's pretty, it looks, looks decent but none of it works. I mean so I'm not sure. It is still a bit hard to figure out what the actual use case would, would be for it right now. Like because if I'm a developer maybe I would just, I wouldn't do this. You know, I would just use my own templates and stuff and get started that way. Or use Cursor probably if you are somebody who's non technical. Yeah, you could use this but if it doesn't actually work now, I mean, the thing you showed, like those simple, like, Python app. That's cool. Like, if you can actually make stuff like that for you, that actually works.

Seems like they all sort of still struggle with creating really decent UIs. Here's what I think each of these things are good for, right? I actually think Bolt is really, really good for creating simple apps. Like you can see on my screen here, this was a prompt. Create a simple app that converts WEBP files to JPEGs, right? One prompt didn't have to prompt it again, and. And it created it, right? Webp to JPEG converter. You drag and drop your files right into here. It'll convert the webp files into JPEGs, right? If you need a really, really simple app for your business to just sort of solve a bottleneck that you're running into, you just have like, a quick idea of something that a pretty simple script can solve for you. Bolt seems really, really good for that. Replit AI also seems like it'd probably be pretty good for that.

We only tried Replit AI on this episode to try to do, like, a website design. We didn't really try it for a basic app. But I bet you Replit would probably be just as good, if not a little bit better than Bolt at creating those apps, just because it does seem to get your feedback as you go, right? It was. You'd probably say, make me an app that creates a web, converts a Webp file to a jpeg. And it would probably say, you know, do you want it to, you know, be able to select where it saves to? Do you want it to be colorful? Do you want, like. It'd probably give you some, like, extra prompts that you can answer the question so that it. It gets more details. So something like Bolt and the Replit AI would probably be really, really good for that. When it comes to, like, creating user interfaces, I feel like that's sort of what V0 was more designed for, right? I feel like when I first heard about V0, like almost a year ago, that's what they were really sort of promoting it as. They were promoting it as, like, a really good tool to create some, like, solid user interfaces. And of all the tools that we used, that was the one that it was clean. It was really simple. You said it had some, like, animated graphics when you first loaded the page, like, that was the one that seemed to sort of nail a UI right out of the box. The very first time you tried the UI looked pretty decent, right? Out of the box.

And the creators of V0vercel, they're also the creators of Next JS, which is like become the most popular framework for creating apps, web apps. So I think where V0 probably would be the best is if you are a developer and you're creating a new website, this is probably a great way to like jumpstart the website's development, right? Versus like people before might just have a template, but sometimes a template doesn't fit what you're trying to do now. So instead of doing that, maybe this replaces having your own template. You go in there, describe what you're going to build, have it build the basic frameworks and the basic layout and make it look decent. And then you go in there and actually add the content, make it do what you want it to do. Replit AI agent, even though it probably had the worst design. But I do see like possibly the most potential in how it's doing things in terms of you've got this AI chatbot that you keep going back and forth with. So instead of having to go code it yourself with like cursor or whatever, like you really just kind of go back and forth like, here's the project here, here's. And then I love it. Ask you what features you want. Yeah, yeah, it didn't work perfectly, but I did do this with, with my son and it did actually work where you could like login. It created a database and you had like a login system and everything. I think eventually for like more serious projects, I think the, the replit way is, is is interesting way to approach this, where you got a project manager you chat with and make sure you're getting what you want and go back and forth. And then you eventually land on, you know, making sure you've got all the features you need.

And then web sim. I'll just say one thing about web sim. Web sim I feel like is really good is if for like maybe you have a domain name that you don't know what to do with, go plug the web, the domain name in and let it figure out what the site's going to be for you. Because that's where websim I think like really shines is just like being that sort of creative brain for you, right. I type in lore.com it has no idea what your lore.com is, but it went and created a site about Middle Earth and Westeros and the continent and the wizarding world. And apparently I can search realms and creators and eras. And so it gave me an idea of what Lore.com could be. So if I have a domain, I'm like, I don't know what to do with this. This is a nice little idea generator that'll give me a little mock up, a little concept that I can use. All of these, let you download the code, all of them. You know, you can go and put it on GitHub, you can download it to your computer, you can, you know, go and do whatever you want with the code once it's created. But they all seem to have like their little pros and cons.

This is already so much better than a year ago. So I, I do believe that when you combine that with the fact that the underlying models are getting better, like the reasoning models and things like that coming out from OpenAI, that probably in a year these are going to be very, very good. It's going to surprise people. And so if you're in business, I would be thinking about what that means in terms of the opportunities. What are you going to be able to do with this new technology once it gets that good? And also in terms of defending your business, what does it mean when people can copy your website or whatever?

Yeah, I mean, you know, from, from a business perspective, I think probably any of these tools are going to be fairly decent at creating like little mini web apps that you can get people to, you know, opt in for if you're just kind of giving away access for free. So, you know, that's, I, that's where I feel like the, the state of AI is. I think if I'm being honest right now, I'm probably still going to use cursor more than anything else. I still really like the, this sort of VS code fork. I like that look, it's familiar to me. I kind of understand it. Like, I'm probably still going to use cursor most of the time. But for people that, you know, they don't want to see all that code. They don't want, like they just want to type a prompt and have the code written for them. These are some alternatives that are out there that you can go play with. They're getting better and better and better, like practically by the day, you know, by the time this episode is even live, they're probably better than what we just showed off in this episode. So it is still pretty impressive. And I think, you know, a year from now these apps are going to be like absolutely mind blowing. Like you probably will be able to prompt a whole SaaS company into existence. Right. Like, I think that's probably going to happen by like next year. So really, really exciting advancements. Really fun to play with. I mean, like, you can lose yourself for hours just playing with these and going, that didn't work. Fix it. And then eventually kind of get it. And there's something fun about that. Like, I don't know why, but like creating a script and getting it to generate something for me and then going, it didn't work. Maybe we try this. That worked, but also now I want to add this feature and just sort of having these conversations back and forth and seeing like a concept that you had in your mind flesh out into something that's actually usable and like, it's addicting. It is really, really, really fun.

Well, that's always been the most exciting thing about coding that most non coders don't realize. Like, yeah, that is the magic of, like, you got something in your head. Now you go code it and then you make it real. And for a lot of people who are not super obsessed with coding, the whole part in the middle is not actually the interesting part. The part that's interesting is like, idea. It's real. That's the part that's super addicting. And so many people have never experienced that. So I think it's. That's a big unlock for people is like, even people who have no technical skills soon will be able to like, go through that kind of creative process of like.

Yeah.

And. And you'll even be able to do the designs and stuff. Right. Like, the design part will get baked in this too. Like, okay, let's flesh out, like, the color scheme. Like you can imagine, like one way. Like, let's figure out your color scheme first. Like, let's. Okay. You like these kind of pictures. Okay, cool. Here's your color scheme. Here's like your five colors, main color, secondary, all this kind of stuff. Accent. Yeah, that's where this is all going to go where you're like, it's going to be entire creative process and you'll be able to go through the entire thing with the help of AI and make things how you want to. And that's. That's so exciting.

So really, really exciting stuff. We're going to be experimenting with more tools. Maybe a year from now, we'll do another episode where we play with these exact same tools and literally see how far they've come. Because they will probably be unrecognizable by then. So we'll have to revisit that in a little bit. But for this episode, I think we'll go ahead and close this one out. Thank you so much for tuning in. Make sure you subscribe to us wherever you listen to podcasts. This one was a very visual episode, so make sure you check it out on YouTube if you're listening to the audio, because we were showing a lot on our screen. But also make sure you subscribe to us wherever you listen to audio. We're on all of the audio platforms. Thank you so much for tuning in. We'll see you in the next episode.

Sam.