Food Bridge

Inspiration

Food waste is a pressing global issue, with millions of tons of edible food discarded while countless individuals go hungry. We wanted to create a solution that bridges this gap by enabling people to donate excess food to nearby community centers in need. Our goal was not just to reduce food waste but also to ensure that those in need receive nutritious meals.

What it does

Web Client for Community centres

Community Centre Registration:
- Community centers can sign up on our application and list their meal requirements.
Broadcasting Meal Requirements:
- Centers can specify meal details, including metadata, time, and date, ensuring accurate food distribution.
Managing Requests:
- Centers can view incoming food donation requests, review nutritional information, and approve contributions.

Android App (Mobile Client for End Users)

Food Donation Flow:
- Users can select nearby community kitchens, donate food, and track their contributions.
Reward System:
- Food donors receive reward points for successful donations, which can be redeemed later.

Generative AI (Llama & DeepSeek)

Fetching Community Center Details:
- If no existing requirements are listed from signed-up centers, AI generates relevant details.
Automated Onboarding:
- Extracts essential details from a center’s website to simplify the registration process.
Question Answering for Food Nutrition:
- Users can ask AI about food nutritional value and suitability for donation.

How we built it

Frontend: Built with React.js & Vite, ensuring a fast and responsive UI.
Backend: Developed using FastAPI & MySQL, providing robust data management.
Client for end Users: Android App: Developed using Android Studio & Kotlin for a seamless mobile experience. -Generative AI : Llama & DeepSeek

Challenges we ran into

Interprocess Communication:
- Enabling seamless communication between different machines on the same network was challenging.
- Solution: Used ngrok tunneling to establish secure connections.
Fair & Efficient Reward Mechanism:
- Designing a balanced reward token system for food donors required careful consideration.
Nutritional Content Information:
- Accurately gathering food nutritional data was a challenge.
- Solution: Used Groq for AI-powered nutritional insights and integrated manual feedback from community center admins to refine accuracy.

Accomplishments that we're proud of

We have a working prototype that aligns with our original vision.
Successfully integrated Generative AI to enhance user interactions.
Fully functional data flow, ensuring seamless donation tracking.
Photo Upload Feature: Users can click and send food photos via Firebase Storage for transparency.
Our database is up and running, efficiently storing and retrieving information.
Developed a functional UI that provides an intuitive user experience.

What we learned

Learned how to integrate AI agents with our application to enhance functionality.
Explored various prompting techniques such as ReAct, Chain of Thought with Hint, Multiple Chaining of Thought, and Few-shot Learning with Random Search to improve AI responses.
Understood how to connect multiple machines within a single network for seamless communication and data exchange.