bitebudget | Devpost

Title Page
Tech Stack
Statistics
Login (Powered by Supabase Auth)
Crypto Wallet Connection Page (before linking with phantom)
Crypto Wallet (once connected)
Social (to view transactions of friends)
Profile Page (with dynamic graphs showing spending over weeks, months, and years)
Friends (real-time friend requests, with transaction/budget tracking once accepted!)

Inspiration

Gamified budgeting meets Computer Vision—scan, save, and compete with friends using Bitebudget.

As a team of self-proclaimed “foodies” and college students, we all know the struggle of loving food while barely managing to save money. Every weekend, we’d splurge on that trendy café brunch or share a celebratory meal with friends—only to cringe at our bank statements afterward. In our conversations with fellow students and hackers, we quickly learned that this wasn’t just our problem. Millions of young people across the country are caught between their passion for culinary experiences and the pressure to build financial security. We saw an opportunity to harness social accountability and cutting-edge technology to help our generation (and beyond) strike a balance between enjoying life and saving for the future.

What it does

bitebudget is a peer accountability focused, social money-saving, and transaction tracking app designed specifically for food lovers, who care about financial responsibility. It allows users to set a monthly food spending limit, and with one quick photo of a receipt, parses the restaurant, price, and other data, updating your remaining budget to not only you, but also to your friends—whether you’re dining out or ordering in. Each transaction is shared with a trusted network of friends, complete with real-time dynamic bar and line graphs, reflecting spending habits and updates on your remaining budget. BiteBudget not only helps you stay accountable, but it also gamifies your savings journey by offering social feeds, likes, comments, and leaderboards that rank users based on how well they stick to their budget. In short, it turns everyday spending into a fun, community-driven challenge that encourages smarter financial decisions without sacrificing the joys of good food.

How we built it

We built BiteBudget using a modern, scalable tech stack:

Frontend: React Native to include both our Apple and Android users, TypeScript for strong type checking, and ExpoGo for development.

Backend: Supabase for real-time data management, authentication, and storage, Python + FastAPI to power our computer vision parsing and transaction uploads.

Receipt Parsing: Integrated a computer vision model via the Hugging Face API + Expo Camera to automate transaction entry with just one photo.

API Architecture: Solana for crypto wallet connections, Google Cloud to host our computer vision model, and Docker to containerize all our dependencies.

Challenges we ran into

Integration Complexity: It was all of our first times utilizing computer vision, and integrating it to parse poorly-formatted receipts proved to be extremely difficult! However, with constant collaboration and lots of documentation, we pulled it off!

React Native Newbies: Most of us had little to no mobile experience, and we all had 0 React Native experience. However, we felt that financial responsibility shouldn't be limited to just Apple or Android, so we challenged ourselves to learn!

Accomplishments that we're proud of

Developed a fully functional MVP that integrates advanced technologies like computer vision hosted on the cloud, real-time database listeners/subscriptions, and user authentication.

Created a real-time social feed with interactive features (likes, comments, leaderboards) that promote financial accountability.

Built a scalable architecture that supports rapid iteration and future enhancements.

What we learned

We learned how to integrate, host, and connect to computer vision models on mobile apps!

We learned how to develop aesthetic and functional apps in React Native to accommodate for both Apple and Android!

We learned how to use Supabase to supercharge our development, and secure our data with Row Level Security policies and listen to changes in real-time using subscriptions!

We learned how to use FastAPI + Python in our backend to connect hardware to our database and microservices

What's next for bitebudget

In order to make our savings tracking process even more seamless, we intend to expand upon our computer vision model, and integrate Plaid to automatically track card transactions! Additionally, we aim to publish our app on the App Store, and invest a lot of time and effort into marketing.