My name is David Andre and here is how to build any mobile app with AI. People are making millions of dollars by vibe coding mobile apps that solve a specific problem. This one generates $2 million a month. This app makes around 80K a month. And this one went from 0 to $10,000 MR in just 72 hours. In this video, I'll show you the easiest way to build your own mobile app with AI. So, make sure to watch until the end. Real quick, if you are a back-end developer, then listen up because we are expanding the Vectal team. So, if you want to work on the cutting edge of AI, then make sure to apply. The Google form is linked below the video. So, let's get to building our mobile app. Now, obviously, there's a lot of different vibe coding tools. However, this one called ROR, which I'm going to link below as well. I think it's the easiest one for mobile apps. So, just go to roor.app and type in your first prompt. I've decided to combine two different viral app ideas, a workout tracker and something like Dolingo. So, as you can see, my first prompt is quite general. I don't describe every single feature of the app. However, I give it enough information to build a solid layout. Now, while it's building the app, let me explain the biggest mistake you can make while vibe coding mobile apps. And that is thinking of a solution instead thinking of a problem. So, a lot of you guys, you have a solution in mind. You have a cool idea that you think might go viral. However, it's not solving any real problem. This is a massive mistake that will cost you endless hours. The correct way to approach this is to find a problem that people are willing to pay money for and then reverse engineer multiple different solutions for that problem and choose the one that makes the most sense. This is how we're going to build a mobile app that will actually make you money. All right, let's go back to ROR. Seems like we've got an error. Let's see. So, I'm going to take a screenshot of this error and I'm going to send it back to RO saying, "We have an error. Explain what it is and then fix it." And by the way, make sure that in here underneath the AI model, the brain icon, you have sonet 3.7 selected because the other two options 3.6 and grog, they're good, but they're not as good as coding as cloth 3.7 set. Okay, so the error is solved, but we see a black screen. So let's describe that again. However, now I just see a black screen. C screenshot always attach screenshots for more context. Okay, so as always, it does some reasoning about the issue. This is because it's using cloud 3.7 thinking. But soon we're going to need to upgrade to the paid plan because running cloud 3.7 sonet thinking for free. This by the way applies to any app that has a free plan. For example, in vectal we have a pretty amazing free plan that gives you access to GP4.1 Gro mini as well as unlimited tasks and notes. However, we are fronting those costs. Okay, so we have a build error here. Let's send that to AI. And actually we might need to upgrade soon because we're running out of messages. So ROR has multiple different plans. Junior, middle, senior, scale 1K. For us, junior will be more than enough. So, let's go with that. Okay. So, I have successfully upgraded to the $20 a month plan. There it is. Junior. If you click on your account, you should see your current plan. So, let's go back to our streak workout app. Look, Clo 3.7 Sonet has no chill. It just deleted like 10 files like that. Okay. But, uh, it got our app running, so I'm not going to complain. So this is our first version. I had to fix a bunch of errors. Anytime you get an error, just screenshot it and describe it and it will fix it. So here's our app. Let's click on start work out. So okay, perhaps we haven't implemented. I'm going to say finish implementing the first stage of our app. I'm going to remind it of the prompt. Now while it's implying these changes, let me explain the second biggest mistake that you guys will make, and that is perfectionism. you know, those of you who are actually building something, you're probably taking too long to release. The app is never going to be perfect. You have to get it out there. You know, sure, at the start, you need to build an MVP, but as soon as the app works, as as soon as the UI is connected to the back end and you can use it, this is when you should release. Okay? Don't try to release it when the app is like 100%. It will never be 100%. You'll always find more things to add, things to improve, bugs. If you release it here, however, you can see whether you even get any users because for most ideas are not actually viral ideas. So the sooner you can release, the sooner you can validate if you have product market fit. However, if you keep building and building and building and building for months, years, and then you release and you figure out, oh, nobody actually wants this solution to the problem. You've just wasted two years of your life. However, if you released after two months or even a single month, then you can do way more shots on goal, right? You can try many many apps and the beauty is that it won't take you this amount of time. If your first mobile app takes you a month to release or two months to release, your next one will take 6 weeks, then it will take four weeks, then 3 weeks, then two weeks and you can just release a new app every two weeks and sooner or later you will find an idea that people are worth paying for. So don't be perfectionistic. If you have an idea, set a hard deadline. 30 days, you have to finish it the MVP in 30 days. then release it and then see if people want it or not. If not, it's probably time to move on. Let's go back to work over here. Sonet 3.7 has applied our changes. Oh, there it is. Finally. So, once again, I had the black screen error. If you have that, just describe it. You know, say that the changes you implemented are causing the black screen. And yeah, so beautiful. We have the zero day streak and at the bottom we have the different categories like our history, our settings, and then two other things that are not implemented. But uh yeah, we have the screen. Let's see what happens when we click start workout. Okay. So, currently just adds to the streak. Okay. Yeah, we haven't implemented that yet, which we're going to do right now. So, here's my prompt for the streak logic. Let me send it. Increment the streak each time a workout is logged on a new calendar day. Reset to zero if no workout logged by 2359 local time. Make the flame glow brighter as the number rises. So, now is implementing the second prompt. Now, while that's going, let me show you this test on your phone feature. Let me move my screen. Actually, my camera I mean so ROR makes it super easy to test whatever app you're building directly on your phone. So this is a big QR code. So just scan that and first you need to download the Expo Go app. So this is the app you need to download. It's called Expo Go. So make sure you get this on your phone. And again they give you instructions for both iOS and Android. So doesn't matter which you have. So just open up App Store on your phone. Type in Expo Go. Load up this app. Once you find this app, just install it and then click open. Next, we need to scan the QR code that we get from Ror. So, this big QR code, open your camera, you know, open your scanning feature on your phone and this will load your app on your phone. See, it's loading and we have a build error. Interesting. Okay, let's actually wait until until this is finished. So, I'm going to click on send to AI. So, now Ror is fixing the Zustand compatibility issue. That way we can run it on our phone. Now when you get these errors, just click on reloadjs. There's a little button, you know, on the phone. Reload JS. Click on that. Oh, and another error. Actually, let's once again, let me scan the QR code. Boom. Opening project. And we have an error. Okay. So, after battling errors for a few minutes, we got the next version up and running. So, let me click on reload.js on my phone. And hopefully we can get it running on the iPhone. Let me restart Expo Go. And keep in mind this is the first like major vibe coding app for mobile apps. Oh, there it is. It actually works. Wow. So, even though there's a lot of errors, you have to realize that is the cost of being at the cutting edge of AI mobile app development. Like building web apps is much easier than mobile apps. There is dozens of VIP coding tools for web apps, but Ror is really the first one dedicated just for mobile apps. So even though it is still very buggy, I mean it released like a month ago. So it's, you know, it literally they applied this, they were not perfectionistic. They released it and now they're improving it based on the user feedback. So it will only get better. But this is the cost of being early. There is going to be bugs. It's not going to be smooth, but it is still by far the easiest way to build a mobile app because I literally have this on my phone. I have it on my phone in a matter of what 15 20 minutes. Even though I had to fix a bunch of errors, it's fine. We have it on the phone. So let's test it out. I'm going to click on start workout one day streak. Okay, log today's workout. So, we need to still implement that functionality. But we do have the history here. That's good. Could also be more visual, I guess. Settings doesn't have many options. And we have error boundary and not found. Okay. So, yeah, but now we can now clicking it doesn't uh change it because it's still the same day. So, at least that's good. So, now what I'm going to do is I'm going to send it the next prompt. And this is the live workout screen. Add a live workout page launched by the start workout button. Timer at the top. Plus minus buttons to log sets and reps. Finish button that save session and returns to homepage. So notice what I'm doing, right? Instead of trying to send everything at once in a one giant prompt like most of you would do with, you know, hundreds of words of instructions. I've decided to split it into six stages of the mobile app. And each stage builds on top of the last one. This is much easier for Ror to follow instead of one huge prompt that is like, you know, confusing with lots of info. Just focus on this, you know, focus on executing the first one. And once you're happy with the first one, once this is fully implemented, then move on to the second one. And this is how you should vibe code. Instead of trying to get everything to work in a single shot, which almost never works, split your apps into five, seven, or 10 different stages. and make sure each stage gets executed fully and completely before moving on. This is the way to use AI to build mobile apps. All right, let's see what the ROR has been cooking. So, it's changing four pages. Workout tsx, workout context tsx, app index, and history tsx. So, let me go back on my phone right here. And this is the beauty. We can test it on the phone. So, let me close it up. Let me load this uh QR code again. See, this is the magic. This is the magic part of RO. what they did. You can literally test your app on the phone as you're building it. You don't have to use an emulator on your computer. This is literally a browser. This is a browser tab. Ror is running in a browser tab. So even if you have a completely shitty computer, 15y old laptop with 4 GB of RAM, you can still use Ror and you can test on your phone in real time, which is pretty amazing. So let's see what's happening. Start today's workout. Boom. It opens up this page. Oh, that's amazing. It's tracking my time. That's good. We have a three exercises that are preset. So, push-ups, squats, and sit-ups are preset. So, you can add sets, you can add reps, but we can also add a custom exercise. So, I'm going to add deadlifts. Okay, deadlifts. We've added deadlifts. So, we have a custom exercise. So, we can do first set. And we can track how many reps we did. Let's do finish workout. Boom. And then we should be able to see it in our history. There you go. History. You can click on first day, May 1st. Okay. You actually can't click on it. Okay. So, the history tracks only the amount of exercises and sets, but you cannot see the workout. Okay, nice. Very nice. That's a huge update. Now, you might think the app already looks pretty good. However, we're only getting started. So, my next prompt is going to be history list. Implement history tab. Scrollable cards with date, duration, total volume, and the streak number captured that day. Most recent session pinned at the top. Now, you might be thinking, "But David, can I really make money with VIP coding apps or with AI startups?" Well, let me show you something. So, in the new society, just yesterday, one of our members sold his MVP, the minimal viable product of his app for $90,000. So, if any of you are wondering, can I make money with AI apps or with VIP coding apps or with building my own AI startup, this post by John should answer your question. So, if you want to start your own AI startup or if you just want to start making money with AI, make sure to join the new society because in the classroom, we have step-by-step modules on how to build your first AI agent as well as how I built my own AI startup from scratch. From literally choosing the video idea to setting up the project in cursor, the back end in Python all the way to crossing the $10,000 a month mark with my startup. So, if you're serious about AI and if you want to make some real money like John, one of our members, make sure to join the new society. It's going to be linked below the video. All right, let's go back to the see what it did. All right, so did it implement the history list. Let's see. Actually, you know what? Let me test it on my phone. All right, so let's see whether the history tab has been implemented or not. There is some code inspector potential issue, but uh hopefully it works regardless. So, I'm going to click on history. So, obviously the workouts are not saved because we refreshed the entire app. So, let's start workout. Timer is going. Uh, let's do some push-ups. Let's do some squats. Two set of squats of eight reps. Whatever. Finish workout. 8 second workout. World's fastest workout ever. And then I'm going to go to history tab. Here we can see it. Okay. So, we see 0 minutes. That's good. That wasn't there before. And we see the volume. Can we click on that? No. Okay. But it already it already did all of that. Let's Let's do one more workout. So, this is going to be just sit-ups only. 10 sets of 10 reps. Just brutal ab workout. Finish workout. Let's go back to history. All right. So, it works, but it marked it as day two even though it's the same day. That's um not optimal, but yeah, our history tab is much better now. We see the duration of minutes and we see the volume. Not bad at all. But let's take it one step further. My next prompt is this street calendar and stats. Inside history, add a stats toggle showing a mini monthly calendar. Fill workout days green and display current stre length above it. Include a weekly volume line chart below. All right, let's see if this is implemented. So, inside of history tab, we should see a stats toggle. So, first let's add some workouts, you know, eight sets of three push-ups. Finish workout. Boom. There it is. We see some stats. Okay. Calendar. Very nice. And we see that wait workout day today is Thursday. Or maybe this is a different time zone. But the calendar works. That's good. I don't know why tomorrow is marked as workout day when today is the is the first. Yeah, first day May 1st. Okay. But yeah, this is a nice very nice stats. Also, look at the UI. Like with with like basically zero UI instructions, Ror is able to create very nice mobile friendly UI. This is not bad at all, guys. Actually, this is kind of impressive. Not not going to lie. Even though there's a lot of errors and bugs you have to fix, once you get it working, this app is is nice. It's nice. I really think this is the easiest way to build mobile apps in 2025. I mean, I literally just speak in plain English. I don't have to do any Cotlin, any Swift, nothing like that. Any Java, any C, plain English, and I get a working mobile app that I can test on my phone. Wow. But once again, let's take it a step further. I want to add visuals that gamify the app because remember, I'm building like a combination of Dualingo and a workout app because Dualingo is amazing retention because of how gamified it is. So, this is my prompt. Gamify visuals trigger full screen confetti and a 7-day streak badge modal when streak hits multiples of seven. After each finish, show a random oneline tip toast at the bottom. Add light dark theme switch in settings. Now, while Ror is building this, I want to say that this summer I'm doing a world tour. So, if you want to meet me in person, whether you live in Europe, Asia, or America, make sure to follow me on IG and send me a message because I'm going to be visiting dozens of cities this summer. So, if you want to meet up in person, make sure to follow me. The link to my Instagram is going to be below or just type in David Andre one. All right, let's see if Ror has been able to pull off this gamification features. Seems like it's still building. Now, another concept I need to explain is the importance of persistence. Most of people that click on this video are not going to watch until this point. So, if you're already here, you're probably in the top 15% realistically, right? But even from the top 15%, most people are not going to build anything. They're just watchers, right? So, don't be a watcher, be a doer. that puts you easily in the top 1.5% you know 10% or 15. But the issue is that most of you who do build something you're probably going to quit within a few months right because building an MVP vibe coding an MVP anybody can do that. I mean all you need to do is to literally speak English. The issue comes after that right after you have your app but you don't have any users or maybe you do have users but they're not upgrading to the paid plan. This is going to be the real test and this is going to be like the 0.1% of people, those who really stick through it. But that's why the payoff is so great. That's why there are these apps that make millions per month or even like 80k a month, right? That's life-changing amount for most people. However, what it takes to get there, most people are simply not willing to put that up. So, if you do want to be in the top 0.1%, those people like John for New Society who actually make serious money with AI, you cannot quit at the first sign of errors. I mean, just in this video, I solved like dozens of errors myself. And I don't take them personally. I just go through them, you know, describe what is happening, take a screenshot, provide ro the screenshot, and continue. With vectal, it took me three and a half months to build an MVP and launch it, let alone to get any payments. At the start, even with my audience, I wasn't able to make money. Think about that. Even with my audience, I was struggling to get paid users. After I fixed some issues, after I improved the landing page, after I added some viral AI features that the users have requested, after I learned all those mistakes, then I was able to scale it to 10K a month and beyond. And again, all of that is documented inside of the new society in the classroom right here. So, you don't have to make the same mistakes yourself. So, my advice to you is be patient, be persistent. Your first app is probably not going to make you a millionaire. Keep going, keep trying. It's never been easier to start a business and it's never been easier to build a mobile app and make money with it. However, you still have to not give up because if you give up, the money will never come into your bank account. All right, let's check on rog and we have an error. See, I'm talking about errors and not giving up on errors. But yeah, so this is just a uncoled exception. So let's click on send to AI. And again, you don't even have to understand what the errors are. If you have persistence and if you use chat GBT, cursor, vectal, roor, all of these AI tools at your fingertips. If you actually utilize them, you will get much farther than you would otherwise. And let me remind you that when I built Vectal, I had zero startup experience. Literally zero. This is my first AI startup. So, it's not like I was a fullstack web developer. I didn't know anything about Nex.js. I was a complete noob. With these amazing AI tools, I was able to go through all these mistakes, implement authentication, implement payments, get my first paid user, solve some security issues, all of that. I was able to do it because I didn't give up and I utilized the AI tools. Okay, so it's using colors directly instead of using the theme. By the way, the best thing you can do when you're, you know, fixing errors or when you're vip coding is trying to learn as much as possible. This is another mistake which I could explain in TJ draw but a lot of you guys when you're vip coding something you're turning off your brain. That's a major mistake. That's a massive mistake. Try to learn as much as possible. Actually the fastest way to learn is using AI tools and trying to build something. Sure you can watch like a 20our course on Python on YouTube but you're going to learn way more if you actually spend 20 hours building with Python. So that's just an example, right? But the same applies for mobile apps. So try to learn as much as possible. Pay attention when there is an error. What was the error and how did the AI fix it? If you actually try your best and if you activate your brain instead of blindly vibe coding, you're going to be in the top 10% of VIP coders instantly because you can actually vibe learn. You can do vibe learning while you're viing. And that's really the way to go. All right, so let's see if we have the oneliner after finish. Okay, so the UI is actually improved because in the settings we have the toggle. Wow, very nice. This is very nice, guys. Light mode, system, dark mode. Okay, so we have dark mode in our app. That's clean. That's clean. Okay, let's start workout. This is also improved UI. Very nice. Let's do squats. Oh, we have squats. Okay, I'm going to do pull-ups plus. Okay, three sets of 10. Finish workout. And there's tip. Okay, that was a bit too fast, but we did get the tip and it should be added to our stats. Yeah, there it is. Volume. So, obviously, we could make further adjustments like showing which exercises this was. Maybe even like running an AI on it to create like a one paragraph summary of what this workout was and what the AI thinks about it. But yeah, not bad. We certainly have some gamification features. And this alone could be an app that could be used to track your workouts. And again, if I spent a few more hours on this and added a bunch more features, you know, polished it, improved some of the UI and UX, this would be ready to be released. And you could charge like $5 a month for this, $7 a month for this, and you would have paying users if you targeted the right audience, obviously. But again, the main thing is solving a real problem. This was just a random idea I chose. But if you're going to commit months to something, you should dedicate at least a month to choosing the right idea. And I know it sounds boring, but that's what I did before choosing the idea for vectal. I just decided, okay, I want to build an AI startup. So, I took 30 days. I wrote down as many different problems and then solutions to those problems as possible. And at the end of those 30 days, I chose the best one. So, that's what I would recommend you to do the same. So, there you have it. Now, you know how to vip code any mobile app just by typing in plain English. So, hopefully you found this video useful and hopefully you take action and you actually start building something. With that being said, thank you guys for watching and have a wonderful, productive week.