I wanted to keep this rather short than long, so if anyone reads it - it actually provides some value and isn't just "AI slop" thats useless.
What it does
Based on Abbey Sharp's dietian views, the app functions as a "combo _ generator _" but I didn't want to stop at an AI wrapper, because anyone can take a photo and send it to ChatGPT, this app offers users an intuitive design, gamification suitable for the target audience (moms), details like sharing your meal transformations (which can also help with marketing),
or the primary feature of saving your favourite combos, so when you're craving chips again, you can simply view what you already like (and STAR your best combos) without needing to take a photo, saved combos offer a "What you need" part and a simple recipe below it.
Challenges we ran into
Some funny challenges were mainly stupid AI suggestions, it just suggested things that tasted really bad, for that (and at the same time optimizing for margins and not using frontier models) I ran a few experiments comparing results across different photos and seeing where the results make the most sense and the price is also good.
Accomplishments that I'm proud of
Honestly getting the app out, I'm yet to be proud of making money from it, but I've never made an mobile app before and I think it turned out fairly well for the amount of experience I have, I enjoy doing new things and for example X (twitter) was helpful, seing some advice or designs that made sense to me and I could implement.
Built With
- revenuecat
- swift
Log in or sign up for Devpost to join the conversation.