BruinBelly

Beli but Better. Beli for Bruins. #1 Dining and #1 Community.

Comment

Inspiration

When we first came to LA, we didn't know anyone. We ate alone constantly. What if we could change that?

What it does

BruinBelly is a full-stack React + TypeScript application that aggregates a 15,000+ entry web-scraped UCLA dining dataset into a structured SQLite backend, enabling real-time menu queries, user ratings, and persistent taste profiles through authenticated sessions.

How we built it

Frontend consists of React, TypeScript, and TailwindCSS. The backend consists of an SQLite database with a Flask/SQLAlchemy abstraction layer. We used a Flask API and Cloudinary to store the short clips and maintain scalability.

Challenges we ran into

UCLA API no longer is public access, scraping took a long time. Connecting the back and front layers was a struggle.

Accomplishments that we're proud of

Integrating fully working database and scraped data with custom ingestion algorithms. Calendar is fully automated with notifications and has a G-Cal export feature. Fully designed within 12 hours.

What we learned

We utilized new technologies like Cloudinary to store all the videos and connecting all the stacks together.

What's next for BruinBelly

  • Hinge-like compatability features to find friends
  • Infinite scroll for short form content
  • Stronger suggestion algorithms.
  • Expanding to web
Share this project:

Updates