I had AI build a mobile app — did it work, how much did it cost, and what did I learn?

The store rejections are likely because you didn't properly declare the camera usage reason in the manifest and missed 64-bit build requirements. Also, AI often suggests deprecated packages, so you have to manually update versions in the pubspec file.

Don't forget the hidden costs here: 1) One-time $25 fee for a Google Play developer account, 2) $99/year for an Apple developer account, 3) Hundreds of hours of your own labor spent fighting changing prompts.

still props for the hustle seriously impressive. getting an app to read barcodes from a camera with zero coding background is a big win. when you eventually hire a developer, youll know exactly what you want and have a working mockup ready, which easily cuts your dev costs in half.

Cleaning up code written by AI sometimes takes longer than writing it from scratch. No matter which developer you show it to, they'll just say "who wrote this spaghetti code?" and want to start over. That 4.200 TL and 3 weeks you spent might have gone completely down the drain.

wish u just distributed it internally as an apk without putting it on the store at all you couldve just manually installed it on the warehouse staffs phones and avoided the whole approval headache

If the app is strictly for your own warehouse staff, stop trying to upload it to the store immediately. Just sign it locally on Android and sideload it directly onto the devices, and look into enterprise distribution options for iOS. That will put an end to the store headache.

When coding with AI, you should go module by module instead of asking for the entire app all at once. Have it write just the data model first, then just the camera UI, and at each step, ask the AI to double-check: "Are there any memory leaks in this code, does it follow the latest best practices?"

I agree, and I'd like to emphasize that. Everything goes well for the first three months; problems arise in the fourth.

Pausing growth to breathe is also a strategy, nobody says that. Proven by experience.

Quick summary for newcomers: When the closure decision was made is as decisive as the decision itself.

When we decide without measuring, we always end up in the same place. Hope this helps.

There is something to watch out for. If it's your first time, start small; scaling comes later.

Of course, it varies if your situation is different.

You're right. Just because everyone does it doesn't mean it's right.

Correct me if I'm wrong.

this thread is archived and taing notes for two weeks yields better results than a six-month estimate.

if it's your first time start small; scaling comes later. i mean i'm also curious if anyone does it differently.

The cheap-looking path usually ends up costing more later. The answer varies greatly by industry; there is no one-size-fits-all rule.

If you have questions write them; I'll answer as best I can.

There's a part I don't understand. Everyone rushing into ai mobile app development gets stuck at the same point.

People defend habits, not processes. Resistance comes from there. I'm also curious if anyone does it differently.

Thanks, this was very helpful. Second-time founders don't make the same mistakes but find new ones.

Hope this helps.

I'd appreciate it if you shared the outcome.

Absolutely. If I were to add anything: Any unwritten clause becomes a point of disagreement later, as both sides remember it differently.

If I were you, I'd go this route.

I went through the same thing two years ago. Mistaking revenue for cash is the most expensive illusion; trust the bank account.

Hasty decisions become decisions you have to fix six months later. This is my opinion, I'm not claiming it's absolute truth.