HomeBites

Inspiration

My parents have always told me to never stay hungry, to eat out if you have to or feel lazy. However, eating out every 2 days sucks according to me and my friends. In addition, when I do cook at home I realize that some ingredients have expired which means I have to throw them away. Basically, many students lack access to affordable, home-cooked meals. I wanted to create a solution that connects students through food, reduces waste, and encourage a sustainable campus community.

What it does

HomeBites connects students who want to buy, sell, or cook home-cooked meals. Users can browse meals nearby, schedule pickups, and even “hire a chef” by uploading their pantry ingredients. AI suggests dishes based on available ingredients, helping to prevent unused food from going to waste. Ratings and tags ensure quality and trust.

How we built it

I used SwiftUI for the frontend to create a clean, multi-role interface. The backend is built with FastAPI and PostgreSQL hosted on Google Cloud SQL, with endpoints for user authentication, dish management, orders, and AI-driven recipe recommendations. The AI agent uses pantry images or ingredient lists to suggest meals, adding personalization and sustainability to the cooking experience.

Challenges we ran into

Connecting the postgresql database hosted on Google Cloud SQL and the Vertex AI API to the UI was hard. Due to the time constraint I could not fully integrate the postgresql database and the AI api with the user interface.

Accomplishments that we're proud of

I'm proud about the fact that I picked up on the Google Cloud interface fairly quickly, which was aided by the first Google workshop hosted at 10:30 am. I better understood how much larger corporations use services like Google Cloud and AWS to scale their projects and databases.