Build it in public(Launching an App with AI)

what's up everybody um i've been building some stuff and just want to jump on a live stream and hang out and talk answer questions uh have fun since uh my twitter account got hacked recently and i feel like i was in purgatory hell for a decent amount of time um but i've just been like going crazy on reassessing design workflows development workflows how i work with clients when it comes to those things how i work on personal projects when it comes to those things and personal projects came to mind so last year i launched this uh this app into the app store called 60 strong it was an app that was basically built almost entirely for my church community so we wanted to do almost like a 75 hard challenge but this 75 hard challenge uh was 60 days long and it included a couple of things working out eating healthy every day praying every day reading scripture every day and i mean we could have just done this with like a pdf and you know hand this out and check this off or printed out a booklet, but I just thought it would be a cool opportunity to create something that was kind of interesting and fun. So here's the app actually being mirrored on my device here. So you can actually find it. Let's see, open it up. It's called 60 Strong. And the thing about it was you could only really log into this with like a key. You had to have like a specific code to enter in because challenge would be set up for an individual community you'd have to have a code or QR code to enter into that community but then you'd be able to basically choose from what type of challenge you wanted to engage in did you want to be like a normal challenge or did you want it to be a let's see if we can get that to work maybe not did you want to be normal challenge did you want to be like a hardcore challenge it's not working on my mirroring device but that's fine And the idea was that you would come in and set, you know, what type of workouts you wanted to do. Maybe it was like you weight trained or did calisthenics or whatever. And you basically just have a track and mark that off. Like, okay, cool. We did it. But then you also got automatically put onto a team. So there was, I think we had something like 140 people sign up. And out of all those 140 people, they got. randomly generated and entered into these teams and I could come into that team and I could have like a conversation with my team so that was one of the habits that you had to do every day was like you know message your team so you would click here and go ahead and chat with your group and it would just take you there so there's like live chat that was happening in the app pretty cool and then there was like a leaderboard for people in the app so you could see who was doing what and obviously click on their profiles so this is The app is kind of dead at this point because... We used it solely for that purpose and it served its purpose and we haven't used it since. And that app was actually developed and built using Flutter Flow. So let me just drag a few of these things around. But it was built using Flutter Flow. So if you don't know what Flutter Flow is, it's like the visual building tool for Flutter code bases that basically pushed this to iOS and Android devices both. So, you know, we were able to... uh go through and build all the individual pages here like for instance like you know the new habit screen and you can see all this is like raw data that's happening inside of there so this was something that i was not incredibly experienced with i got to do a little bit of it myself obviously we started all this from uh designing everything in figma you can see all the different flows and everything that were custom designed so light mode dark mode and you know adding your habits and all this kind of stuff and it was great this took me a while to do took me a couple weeks as a personal project to kind of crank that out and then i started working inside of flutter and it's kind of there's a there's a big learning curve when you're working inside of flutter flow or any sort of whizzy wig kind of building tool I basically had to onboard a Flutter Flow developer that helped me. That took another month and a half or two months. So all in, it took me about three and a half or four months to get that app into the app stores. And it's still very much in the app stores right now, both the iOS app store and the Google Play store. So you could find it there. You could download it. You won't be able to do anything with it, right? Because you need that code to get in. I wanted to relaunch, I always wanted to relaunch this app Because there's a couple problems with it. Obviously, I've already mentioned it was very team based. You needed a code to get in. And there was just like a lot of it was only ever meant to serve that single purpose. So I had always hoped to rebrand it and turn it into a single player mode daily habit tracking app built specifically for Christians. I'm a Christian. I'm a person of faith. So for me, what's important in my daily habits is to not only do the things like the working out and the, you know, the taking. my vitamins and all that kind of stuff, but then also to be reading scripture and be praying. So I wanted to rebuild this app and relaunch it with a few kind of changes. A, it's single player mode. You can just log in, sign up and use it individually. B, I wanted it still to have the mixture of daily habits as well as reading scripture. One of the things that was really, really difficult to do was to get scripture into the previous iteration of this. So, you know, when you, for instance, were going to read scripture, these were prebuilt plans that I manually uploaded into here. It's really janky. You know, it's not a great reading experience. It just kind of sucked all around. So it wasn't a great experience. I wanted you to be able to read. I wanted you to be able to create a habit that's based off of. scripture reading plan. So pick certain books of the Bible and that becomes a plan that's divvied up over the days. And then you can manage that. I still wanted to have that challenge aspect to it and a little bit of the community aspect to it. But one of the biggest things was I didn't want to spend another three months or four months. That's, you know, fast forward or rewind back, you know, four or five years. That's an insane pace to launch an app to the app stores in three months. Nowadays, it's just, that's far too long. We're just super spoiled and privileged now. I don't want to spend three or four months doing it. So I, you know, I've been diving into Claude Code. I've been using Replit. It's probably one of my favorite vibe coding tools right now is Replit. And reworking things. Recently, I saw that Replit released a new Expo Go feature. Basically, you can publish directly to the app stores, leveraging another tool called Expo Go. So I've gone ahead and done that. And you can see I just spent some time. prompting it probably took me i'll be honest i ran through my monthly credits at like whatever it is 30 or 40 bucks a month and i spent another probably 150 dollars in credit usage to prompt it into a place where i like it so um i'm gonna open it up and kind of preview it and show it to you i fed it some basic kind of things but what's really cool is that replit allowed me to rebuild the project and although it it uh it's it's built in react native so it's going to adapt to the platforms based off of which app store gets loaded into which is super cool so let me re-mirror here and go up and you can see uh you know i can actually build here in replit as long as the app is running i should be able to uh actually open it up in expo go which you're seeing now so let me just kind of clear some things out of the way and zoom in so here's What about $150 worth of prompting got me was the revised version. You can see it has the date picker at the top and it's kind of hard to tell maybe on stream, but we are getting a little bit of that like kind of liquid glass effect. So it's kind of cool. So you can kind of scan through even when you open up the, you know, the settings and everything that's working fine. And I can. Like any basic habit tracking app, I can mark off those habits and it's tracking my progress here, which is pretty cool. So that's nice. Also, right out of the box, you know, you got the ability to do light mode, dark mode. So it, you know, switches over, feels really nice. That's good. Match my system, all that kind of stuff. And so let's apples to apples. It's doing everything I want it to do. What you couldn't do before was have a scripture reading plan. So I'm actually going to delete this one that I have on. the uh the board there and we get like a little bit of a marketing banner here to join a challenge or do this but you can basically go up to plus and start a new scripture reading plan and this is really cool this is actually kind of complex functionality the first thing you should know is that uh i wanted to be able to read the bible no matter what the person wanted to read so i basically created a bible reading interface tapping into bible api which is when you're just in a sandbox environment 100% free. So now we're tapping into it. We get like a nice, real simple reading structure here. I can move around between Old Testament, New Testament, select my chapter, I'm reading. And then to kind of comp other Bible reading apps that are out there. I wanted to have some basic reading functionality. Like, for instance, you're able to kind of navigate around through the chapters. All this is just being grabbed through the API and updating everything. But I also wanted to, you know, on read, my main navigation goes away, which is kind of nice, on scroll. But also you can click on... Text inside of here notice like I get this nice little underline here And I can do things like highlight scripture or I can leave a note here So for instance, let's click note and write a note like you know, this is cool Save that note and now we get like the ability to look at that note edit that note and then back in your Account you can look at all of your notes and check them out there. So that's kind of cool Whoops we just Went back. Let's go out of there. Okay, so let's Let's close that it's a little bit weird on mirroring Okay, fine. So, you know, we're back in scripture. You could read scripture and do all that kind of fun stuff Which is great because now it becomes a little bit more sticky. It's like a daily usage application I can come in here and I can read any part of the Bible I want but I can also go back and check off all my habits what's the cool part is now that we're tapped into the bible api which i'll have to pay for that's not a problem you can come in here and set up a new scripture plan and this is where it gets really cool because you basically start to select multiple books of the bible like maybe i want to read these ones great great great and it tells me down at the bottom as i'm adding them how many are selected how many chapters and if i want to read this in maybe two weeks it's going to be about 32 minutes of reading a day or 16 minutes of reading a day And I create that and it's grabbing all that information from the API and then actually separating it up into even reading plans so that every day when I come in, I can see the chapters that are assigned to me. So, you know, today I have to read chapters one through three, scroll through, we get a nicer reading experience, you know, get to the bottom of it and mark that off as read and that counts as your habit. That part right there, actually just building out the basic habits and the interface, I would say probably it only took about 25 minutes or 30 minutes of prompting. Then building the customization on how reading plans work and bringing in all the scripture from the API probably took another hour to kind of figure out and prompt. It's been teaching me a lot about prompt. engineering or just kind of like the way that you prompt um has been super interesting um somebody asked me uh did i use figma for design so actually yes and no so what what i did was i uh i actually brought in old screens from my previous design so we actually i was going through like a phase two of the design uh to eventually get it into that single player mode using the same theme and style so you can see this was like you know it was still real like sporty a little bit masculine you know kind of challenging and whatnot and i was reworking it this way so i yes i brought in these designs and that gave it That gave Replit the context to understand what I was trying to accomplish. I could start dialing it in. But then I just decided I want this to be an application that is not challenge-based, daily use-based, more broad as far as the demographic or people who could log into this. I want moms and teenagers and I want anybody that is a Christian to feel comfortable logging into this, not somebody who's just real challenge-based or aggressive about their habits and goals. So I loaded all of this in. And then I actually just grabbed, let's go back, I'll go back out. I've been using this app called Brick, which basically bricks my phone. and locks it away so that uh it turns off all the social media apps and everything so i break it in the house before i come out to this office and i just really liked the neutral kind of clean style of this application so i just decided i uploaded a couple of examples from this interface and i said restyle it to match now what's you know interesting about ai is uh i've given it the context the functionality what i want the app to actually do I brought in some designs to help give it that context. And for the first couple of hours of prompting, it looked like that old version, that athletic, you know, more masculine version. But then asking it to retweak the theme, it flipped over within minutes, which is really interesting. And from there, there was a lot of extra little tweaking and changing stylistic prompts, you know, about using color. In this version, there is it's void of any sort of color. So I wanted to have some sort of accent color, for instance, for like. progress metrics and, you know, things like that. So, um, and I also brought in a bunch of images that have kind of, I generated with mid journey, a bunch of these kind of, uh, wood grain, um, imprint style images, uh, like, uh, Alexander Dior's old style, like, you know, wood, wood grain print. So I brought those in as a style as well. So, did I use Figma? Yes and no, I did use Figma, but I really didn't spend a lot of time in Figma. I think if I was to start this whole process again, I probably would have some initial screens and the most important screens would be those, the happy journeys or the main flows that I would need and the theme that sits on top of it to match. I'd probably map those out pretty quick, but I wouldn't spend a lot of time designing. And actually, I was just... copying and pasting things back over from Replit into Figma, I would probably design about this much. Like this, there's like four or five screens here of core functionality that, and for me, the biggest thing is describing functionality and giving context. So for instance, if I was to design these from scratch and then feed these into Replit or Claude code, I would then be able to point things out. The horizontal calendar navigation, scrolls and does X, Y, and Z, the, you know, categories of habits like scripture reading here daily and weekly oh what does it look like when you create a habit you know maybe some following screens but once i get the basic core screens and then flows kind of down then i would probably spend the majority of that time inside of repli yeah if you're just joining me we're relaunching an app into the app stores that I currently launched using Flutter Flow, which is like a visual drag and drop Flutter builder. And I'm now redoing it with Vibe Coding Tools. In this case, we're using Replit to do all of this. So yeah, so we've prompted, reprompted everything into existence. And then what's cool is, you know, if you're going to preview it, you just make sure that you start out the project saying, I want to create a... react native project using expo go and it immediately gives you this option so you could actually scan this right now and use the app if you want to just take out your phone and and do a little scan it's going to ask you to install expo go on your phone and i'll show you again what that kind of looks like if we just mirror my phone here you're gonna have expo go and we can well i can't really do it Here can I because I need to like three press or like triple finger press but I can't do that to restart the app but anyways the app is basically running in the production environment in Replit and then it's acting like the staging server as I'm testing and prompting and demoing my app so It'll it has You know authentication database login structure all that kind of good stuff I also created which is kind of fun recreated the whole challenge aspect because that is a still a part of What the old app did So, oh, you have ExpoGo. Yeah, you should be able to scan that code right there and get it, this code that's on the screen on the right-hand side. And so I still wanted to be able to create challenges and then have people explore those challenges. And basically what these challenges will do if I was gonna go ahead and join this gratitude challenge, I hit join, it's gonna tell me, okay, these habits are gonna be added to your daily habit screen for a certain amount of days. So, you know. that these two daily ones in this weekly one if i was going to do something uh and also i had the concept of creating live challenges things that only show up for a specific amount of time and then disappear so let's say we're going to do this like gospel deep dive we click that it's going to tell me a reading plan is going to be implemented on your habits and journal entry and sharing your insights. So if we join that challenge, boom, we can do that. And you can go back to today and you can see those habits have been added. This is coming from the gospel deep dive challenge, your little challenge iconography there. And then now we have these new insights and there's different actions, right? If this is a custom habit, like I've created here, I can long press on it and I can edit or delete those habits. If it comes from a plan, I can either leave that challenge or go to that specific challenge because this has all the details. on that challenge and it also has a little bit of that community aspect so i can come in here you can see i've just been testing this out and say you know doing good today or something and post that so now we have kind of a community aspect to each of the challenges and it's working and so from here we could actually leave the challenge and if we try to leave the challenge it's going to tell you you're going to be all those habits will be removed from your daily or your today screen boom we leave the challenge We're opted out of it. We're right back to our normal challenges. So super interesting the way that it works. And yeah, so habit creation, there's still some things I need to work on here. Like when you create a new habit, obviously you could, you know, walk 2000 steps and then we can give it an icon to match and then select it as either daily or weekly. So maybe I want to do that like three times a week. I go ahead and save. Boom, that's been added to my weekly habits and my progress. So kind of a cool experience to build it. And this is one of the things that's been really interesting is there's been a lot of these vibe coding tools, whether it's Lovable or Bolt or Replit. And the problem for me is always the process of... developing it having it working in staging testing it but then actually getting it into the app stores um and i am a fairly technical person i just don't want to spend the time going from here and taking the full code repo and everything that i have over here which i do have you have your full code repo that's happening i could push that up to github and then work off of github repo get it into into the app stores so on and so forth but the cool thing is that in replit they now have the ability to push uh straight to production and that actually works through my expo go or my expo um account so you can see i have that app here called rhythms and i can connect that to github directly which i definitely am going to do here in a little bit but it also gives me the ability to create builds of this app. And I actually straight from Replit was able to press a button here and say, push it to the app stores, which was super interesting. So now actually when you open up my Apple developer account here, you can see that there's my old app and here's the new version that's coming in right now. And I actually can. Push a bill or create a build of this and send it to test flight. You can see there it is ready to go I actually got the invite on my phone. So let's see if I can get the The invite actually running here and pulled from my phone Because I did get that notification that that was a thing Let me open up test flight and see if it is actually working Oh, it looks like it's not right now. Did I get a text message? No, I forget how it comes to you. I thought it came to me in email form, but I'm apparently incorrect. But anyways, I do think, oh, yep, it came to me by invite. Let's get that. Yep, look at that. Okay, let's mirror my phone again real quickly. see if we can walk through the process and get this i'm kind of just i'm building this in public and to see what happens so let's open it up in test flight holy cow okay so we're installing the actual test flight version of it and it's kind of working let's see holy cow okay so let's actually just let's drop back out really quickly to my phone and see there there it is it's sitting in a test flight build And open it back up. Yep. Okay. So we can move through the onboarding. Great. Fine. I got to fix that onboarding because it super sucks. And then I have a native log or email login, but also sign in with Apple. So let me see if I can sign in with Apple through the test build. Should be able to continue with passcode. Authentication required. That didn't work. Okay. Let's see if we can do it by email and see. See if login works at all. Watch it. Save it. Okay. Yeah. So that's a problem. That is a problem. So yeah, I can't log in. I got to figure out what's happening with the authentication of it. That's like normal stuff. But I mean, the good news was I was able to actually push a test flight build and get it here to get working. Maybe I can sign up for a new account. Maybe just maybe let's go test. let's go test at test.com and we'll do a password and apparently I don't have a button for email signup or login so that's stupid I'm just pressing enter and hoping that works that's something that I need to fix why don't we go back really quickly and fix that inside of replit so I'm gonna go back to my overview here and I think this is this is I mean this is the process right is just prompting it And I'm going to open up my new favorite tool called Whisper because I don't like typing. I'd rather just talk. So let's come here and just say something like, there is no login button for email login or account creation in the login flow. Okay, great. So we whispered that into the prompt and let's see if it goes ahead and fixes it. And while we're doing that, we might as well log out really quick. I do want to just note here something kind of cool on the Expo Go version of this because... I need the ability to manage it. There's always that like that back end piece to creating apps like user management. And for me, I have to create challenges and publish them and unpublish them. And if we're going to have user generated content with comments, people have to be able to flag those. So I needed a way to kind of like manage all of that. So if I open up Expo Go again, you can see that I have an. for me as a super admin and only I can be a super admin you basically open that up and I have I have to reformat this a little bit because it kind of looks like junk but we can open it up and eventually I'll be able to manage users comments here as well as challenges but we could for instance that gratitude challenge that's currently published I could unpublish it and there you go you see that it's like unpublished and let's just bring this back around and see if we can click up here yeah okay no no no little funky but i mean well we also have the ability to create new challenges which is kind of interesting so you can select the different books that you want to add if there's reading portions of it and then add new habits to it as well which is kind of cool and for some reason we can't do that so i might need to close it down and then relaunch it okay so uh oh what was kind of interesting i'm gonna hide that what was interesting was i didn't realize i was in plan mode inside of repli which i totally was and it says like okay here's the plan tasks to check the login flow screens everywhere fix form cutoff issues verify login screen works um we can kind of revise that plan Actually, let's go back and just say, like, yeah, start building. I don't know. Maybe that's not the right plan. But someone said, yeah, your sign up is a little bit messed up. Yeah, it's a little bit messed up. Definitely, I have to fix that. But again, this is part of the testing process, right? Like, that's okay. I can figure that out. If I can figure out, like, all the complex functionality and prompt that into existence, I can surely fix, like, a broken login flow. That's no problem. for me the more exciting thing is to reduce that design and development time from three to four months now down to three or four days and look that i already have a version of it staged inside of the app store so i'd have to do the google play store next but that's pretty impressive to get to that stage i think um so uh let's talk about costs here a little bit because that's been kind of an interesting one um There's a few things from a cost perspective that I'm going to have to pay for. Like I have a Replit account. I pay for that. I'm going to have to move off of the sandbox mode of Bible API onto a paid account of that. That's like 25 bucks a month. That's not bad. I have to set up Revenue Cat. which will help to capture paywalls and subscriptions for the app. So I'll have to set up those paywalls and different pricing tiers. There's going to be a free tier, obviously, and probably a couple bucks a month. If you want to support the app, it's basically to support those monthly costs and the usage costs of everything. William Mayo asks, is Replit free? You can sign up for a free account and there's limited usage. But I'll tell you, I like Replit a whole lot. I think it's pretty stinking great. But Claude Code is also pretty fantastic as well. The other day, I built three apps in one day. I built a web app, a mobile app, and a native Mac app. I'm taking a course. This is not. paid promotion but i'm taking this course and who made this course i found it on gumroad and apparently i can't log into it right now but it was called claude for designers claude code for designers this one i don't know who made it it was like 30 or 40 bucks it was worth it to me um but that helped me get claude code up and running and it has like a series of like nine projects so those were a bunch of the projects that i built yesterday i built yeah this this web app link previewer i built this uh color drop one and then i skipped the habit tracker one to see what i could do with clod code and it was pretty good so i bought this one again this is not i'm not promoting it but i just hey when there's something cool that's helpful to me i figure i share it with you guys that's a pretty decent investment um and uh somebody said is replet free i thought they had a free plan do they not have a free plan let me i'll i'll go back and check i swore that they did have a free plan maybe i am i could be wrong i don't know i've been signed up for the thing is like that the conversation about like free is you're never going to be able to do everything for 100 free that idea in my opinion is stupid and stupid for two reasons um if People complain all the time like figmas like it costs money or framer cost but yeah But like if you're building if you're using figma to make money Then it's an investment in making money if you use framer to build websites for clients and make money Then it's an investment in making money if you use replet to make money Then it's worth the investment to make money if you're not making money and you're just playing That's a different thing altogether, right? I mean you could use Claude for free to a certain extent. I don't know if you'd use Claude code, but all of these things are very, they cost money, but it's only a problem to you if you're not making money or validating the expenditure, the overhead of them, because you don't have a good idea that's going to make you any money. So that's really... that's the idea there you know so yeah they might have a free plan it's bad but is it good enough to get you up and running and understand the platform and figure some things out yeah probably and then saddle up and come up with a good idea and figure out how you're going to monetize on that idea and you go oh no i'm paying 25 bucks a month for the bible api and 25 bucks a month for replit and 20 bucks a month for expo go and And $99 a year for my Apple developer account. Like, boo-hoo. I had to spend $200 to get this app up and running and $100 of monthly overhead costs. $100 a month? $1,200 a year? Like, yeah, I know that maybe seems like a lot if you're not making any money. But if you're making $1,000 a month off of your app, if you're making $2,000 a month off your app, isn't that $100 a month like a validated cost or expenditure? It is. So when people talk about... paying for tools and how everything in the world should be free i'm like shut up dude like figure out how to make some money off of it and Stop whining. Stop crying. So yeah, those are my thoughts. So anyways, kind of interesting. So yeah, there's a couple components there that I still haven't figured out and still need to figure out, which would be, you know, obviously testing it, getting all the bugs out of it for login, sign up, account creation, management, yada, yada, yada. The paywalling using revenue cap. That's another thing. That I'm still setting up and figuring out and whether or not I want to continue to use replit or I could at this point most likely move that code base into a GitHub repo and use Claude code to work on that repo from here on out. I'm personally kind of happy just continuing to work on it inside of Replit because it works really nice. One of the things that actually makes me kind of a fan of Replit is this process right here. Let me just bring this up. which would be to mirror my phone again and show you. I am a big fan of actually building the app, like using the mobile app, right? So like I was at my son's basketball game the other day and while they were like in halftime, I was here inside of the app reprompting and working on a few things and fixing it. So I was able to, you know, make some changes and fix some things here. And it's pretty, it's... pretty stinking great like as it's working it'll just give you those notifications and say you know like there's like there's something happening in replit okay it finished running its prompt so like i would prompt and be like oh you got to fix this little thing or change the color to that or the functionality is not working boom put my phone down watch my son's game it's you know time out or halftime pick back up oh it accomplished something i'm working on it again and so the ability to work on it from the mobile app is pretty impressive pretty cool you could see here like you could do everything here you could preview it you can prompt it you can do your publishing work directly from here i can look at the actual console but you can open up any of the other things you want to look at like do i want to look at oh where did it go where did it go let's go back into my apps open up my app that i was working on But you can open up anything else as well. Like I can open up a new tab and check out like the app storage or the database structure, which check out the database structure. Here's my development and my production database, different databases for each. So if when I, I open that up, you could see like, okay, what's my, what's my like database scheme on structure happening. I'm doing all of this from mobile. Like that's pretty cool. that's that i mean it's just pretty cool to be able to do it on the mobile while i'm out and about like go in the house make some lunch i'm prompting oh it's updated cool do something else make my lunch so that's kind of an interesting process so um my goal here is to get this fully published up to the app stores through testing through beta um and maybe invite some of you guys uh you know to the test flight and have you guys test it out for me Then actually launch this this thing with a couple bucks subscription There is also I haven't I didn't really touch on it or hit it yet But and I'm I'm not even sure if I want to roll with this, but there is also the option That I have played around with the idea of adding a companion app Portion of it like an AI chat so you can see Currently, let's reconnect this. I'm not 100% there. It's more of like, I'm not sure if I want that in the app, but there is a companion tab that you would open up here and then that would hit you with paywall gating. It's a AI companion. so that you could be chatting and talking with the ai asking it questions i have it kind of locked away right now but the idea would also be that notice like when you're reading scripture it's kind of i think it's kind of fun and interesting i can highlight some scripture like this and i get these options to highlight and note but i can also just ask the ai and it'll bring those scripture references into the AI and say, tell me a little bit more about this, or you could say, I don't understand this, explain the culture, the context, so on and so forth. I'm just a little bit sketched out on allowing AI to give biblical or theological references and information. You gotta be kind of careful about that. It says, if somebody signs up to the ProPlan using your referral link, then you both get $10 in free credits. oh i didn't know that i don't really oh i don't know maybe i'll post it there later if you want free credits but i'm not really worrying about that right now that's not the point of the stream it was just to kind of show what i was working on um again so let me let me just kind of clock it really quick before i log off because i think i've spent a total again about 150 of credit usage in inside of Replit. I've spent no other money on these other platforms because I haven't moved over from Sandbox API or RevenueCat or ExpoGo. I haven't upgraded to any of those yet. I probably will need to. But about $150, $175 at this point to build it out the way it's at and get it into a test flight. Time-wise and energy and resources invested into it is probably five to six hours a day for... total of four days. So maybe you're talking about 24 hours total, like max to get everything sorted out. So about 24 hours of my time invested into it, I would expect I'll spend another 20 to 30 hours finessing and hooking everything up to actually get it uploaded to both of the app stores. So you're talking about a total investment of a little more than a 40 hour work week. a couple hundred bucks to make it happen. I will have to spend some more token usage and replet to make it happen. So that's to me the most interesting thing, because if you have an idea, you're no longer, you're no longer even just locked down to, uh, I can do my portion like the design, but I got to hand it off. You're now able to, to materialize all of that, to actually manifest all of that and make it something that's real and functional. Now, what does follow-up support? uh and updates and what does security look like what i don't know yet um that might be something i have to tap on you know uh somebody like my friend andrew daniels out there who who builds mobile apps and does stuff like that and and get somebody to just take a little look see a little little little look see see into and see if it's secure and stable and handle errors and issues but from my so far my experience has been if i have an issue um i can kind of depend on replit for a lot of it because i can come in here for instance they have this feature that allows me to make like fast decisions or do app testing uh at different levels like on low level it's just going to execute on tasks on medium it's going to review code changes and fix issues on high. It's going to expand and review the scope to the entire app. So if I change something in one flow or one particular view, and for some reason that breaks the code base somewhere else, it's going to find that. I don't have to tell it. I don't have to demo the app and go, it's broken, fix it. It's broken, fix it. I just demo the app. I give it instructions. It starts to demo and test. So it goes, we found an issue we're fixing in the background for you. So that's to me the most interesting. So Shubham says, are you going to stream regularly? Yeah, I think I'm going to make this a bit of a live streaming series as I continue building this thing out in public. And what you can expect as I continue to build it out is show you the ups and downs, the pitfalls, the areas that I get stuck, where I get frustrated, where I... I'm slamming my desk out of frustration, but then also invite you into testing to try some things out. So I would imagine at least a once a week stream showing this build in public kind of process until we get it. And what I would love is to get it and then have somebody actually sign up for it and pay a couple bucks a month and we could celebrate that first four dollars made. That would be exciting. And I think that. that would be kind of an interesting journey because again even if this app makes a hundred dollars a month or two hundred dollars a month like all you need is a couple of users to do that um it only took it's only gonna have taken me a handful of hours to do it it's possible right so anyways uh interesting thoughts interesting process i'm really enjoying it as i play around more i think also These are not sponsored or anything. So I'm just going to be really, really honest about things as I am using the tools and learning more about, for instance, like Claude Code and start to share more of my preferences on this process. I was tweeting about this the other day that that old product design process is kind of dead. It's done. And what I'm showing you is what I'm starting to what's starting to materialize for me as a newer. product design process which skips through really long sprints of of prototyping and testing and actually getting idea quick theme or or visualization of things into a working app with data and authentication and then testing it that way. I don't have to worry about, for me, clickable prototypes in Figma are dead. They're completely dead. I never want to do another one in my entire life. If I never had to, I'd be the happiest person ever and I don't think I'm going to because if I can get something here and test it in this way, I mean, just even here inside of Replit, like... testing things and figuring out how the flows work. There's such a big difference between looking at design and looking at flows laid out versus walking through them, experiencing them, feeling them. There's some sort of magic in the ether about how it feels, how it all comes together. And that's... That's for me, one of the reasons to get to this process quicker. So I'm gonna start documenting this a little bit more. Thanks for watching the stream. Hope you guys have a great, was it Wednesday? Yeah, I got some client meetings how to go to. And we will chat soon, I think. Take care, everybody. Have a great day.

I'm trying to relaunch a mobile app project that I published to the app stores last year with new AI techniques. Follow along for the journey.