Recipe Organizer - Food Fit

Splash screen and branding
Main recipe import menu
Homepage and recipe search page

Inspiration

Most of us have recipes scattered everywhere: screenshots, bookmarked blogs, handwritten cards, and videos we swear we'll come back to. On top of that, figuring out whether a recipe actually fits your goals, like eating healthier, usually means plugging ingredients into a separate app. We wanted one place that keeps your recipes organized and tells you how each one fits your lifestyle.

What it does

Food Fit lets users save and organize recipes in one place. Each recipe is automatically broken down into ingredients and cookable steps so users can see the recipe at a glance. Users can import recipes from URLs and scanned from photos, and save recipes to cookbooks. They can also use an AI voice assistant to guide them through the recipe hands-free.

How we built it

We built the frontend with React Native and Typescript and Sqlite. The backend was built with Node.js, Hono, Supabase, Revenue Cat and Clerk. Recipe data is stored in Sqlite on the frontend and Postgres on the backend. We us 3 different AI models for the app. The voice assistant uses Open AI model gpt-realtime-2.1. We use 2 models to digitize scanned recipes, gemini-3.5-flash and gpt-5.4-mini. The app is cross platform as it is published on iOS, Android and Web. The backend is hosted on Render.

Challenges we ran into

We wanted this app to be functional offline, so we optimized for building a local device to cloud syncing system. The challenges from syncing came from managing two different databases (Sqlite for local and Postgres for cloud) and from managing images properly. We also spent a lot of time implementing and refining the voice cooking assistant. There were challenges with getting the audio to play in speaker mode. We also spent a good amount of time refining the UI and having it sync properly with the audio. Lastly, we implemented a Live Activity widget on iOS to send a user notifications from their lock screen. We did a similar, but simpler, widget on the Android notification tray.

Accomplishments that we're proud of

We shipped a cross-platform app (iOS, Android and Web) that allows users to import recipes from a variety of sources automatically, save them to curated cookbooks, and cook those recipes with voice assistance. We did all of this in a month. We are proud of the UI and styling of this app. It feels more refined than a typical v1 product because we also have a paywall with subscriptions, analytics via Posthog and Authentication/Social Auth via Clerk.

What we learned

We have learned a lot of strategies and techniques for using AI to help us build products without compromising quality. We have also learned about our ability to iterate quickly and build something substantial from scratch.

What's next for Food Fit

We would like to add nutrition tracking, meal planning, and social sharing features next. We also have plans to integrate with shopping carts to allow users to send their ingredients lists to their local supermarket. The longterm vision for Food Fit is to become a singular app to help people manage their nutrition.