Bruin BiteShare

Image Processing for accurate descriptions
Shelter dashboard
volunteer mapping and login
NLP for shelter input
Volunteer match

Inspiration

UCLA is known to have the #1 dining hall in the country. However, every single day, UCLA dining halls throw away an astonishing amount of food — roughly 8,700 meals wasted daily. Seeing that abundance go to waste while thousands of people in Los Angeles struggle with hunger moved us to act.

What it does

Bruin BiteShare connects UCLA dining halls with homeless shelters through an AI-powered food rescue platform. It uses image recognition to automatically identify and describe leftover food (including nutrition facts) and natural language processing to understand and match shelter requests written in plain English—eliminating the need for manual data entry. Beyond matching food with those in need, Bruin Biteshare also coordinates the people who make it happen. The platform includes a volunteer matching system that pairs available students and community members with pickup and delivery opportunities based on their schedules, locations, and transportation options.

How we built it

With the help of AWS Bedrock and Amazon Q, we started developing the backend of the project together first. Once we had a solid idea of what needed to be done, we split, each taking up a major feature of the web application and fully developing our respective portion.

Challenges we ran into

Struggled incorporating the NLP, and struggled to connect with the AI agents. We sought help from the AWS employees and presenters for suggestions on which models were best for image recognition. We also struggled to create a dashboard that saves shelter requests, which we had to sit down together and solve.

Accomplishments that we're proud of

Ending up with an accurate image-recognizing feature, as well as being able to incorporate the NLP. Also being able to launch a viable MVP. This was all of our's first hackathon, so we're proud that we finished with a project we're all proud of!

What we learned

We learned how to make a full stack MVP in a day using various several coding languages and tools. We also learned how to use the AWS platform, not just on this project, but truly learned its applications which we all see ourselves using in the future.