Nutrilink

Delivering food where it matters, before it's too late.

Comment

Inspiration

The idea of Nutrilink came from a simple but powerful observation: While millions struggle with hunger, vast amount of perfectly good food are wasted every day. We wanted to build a system that doesn't just list food, but actively ensure it reaches the right people before it expires.

What it does

Nutrilink is designed as a real-time, geo-based platform with strong frontend and backend integration and AI driven insights. It connects NGO with food provider efficiently.

How we built it

NutriLink is designed as a real-time, geo-based platform with strong frontend-backend integration and AI-driven intelligence.

  • Frontend: Built using React + TypeScript Tailwind CSS for a clean, modern UI Framer Motion for smooth animations Google Maps API for real-time visualization

The core interface is a map-first experience, where users can instantly see nearby food availability and interact with it.

  • Backend: Built with Node.js + Express REST APIs for: fetching nearby food listings managing requests handling user interactions Filtering logic (distance, status) is handled server-side

Challenges we ran into

  1. Real-Time Data Handling

Ensuring the UI stays in sync with backend updates without lag or inconsistency.

  1. Map Integration

Replacing static UI with a fully interactive map while maintaining performance and responsiveness.

  1. Meaningful AI Usage

It was challenging to ensure AI:

  • actually improves decision-making
  • is visible and understandable to users
  1. UI Complexity

Early designs became cluttered. We had to:

  • simplify layouts
  • maintain visual hierarchy
  • focus on what truly matters
  1. Frontend–Backend Alignment

Ensuring:

  • no duplicated logic
  • backend remains the single source of truth
  • frontend stays clean and reactive

Accomplishments that we're proud of

  • Built a fully functional, real-time map-based platform that connects food providers and NGOs seamlessly
  • Successfully integrated Featherless AI in a meaningful and visible way (not just as a backend feature)
  • Designed a clean, intuitive UI that handles complex data (maps, filters, dashboards) without overwhelming users
  • Implemented end-to-end flow: listing → discovery → request → tracking
  • Ensured strong frontend–backend alignment, with backend as the single source of truth
  • Transformed a real-world problem into a practical, scalable solution

What We Learned

  • How to design and build a geo-based real-time system from scratch
  • Importance of clear data flow between frontend and backend
  • How to integrate AI in a way that is useful, visible, and explainable
  • Balancing aesthetics with usability — avoiding overdesign while keeping it engaging
  • Working with maps, geospatial calculations, and performance constraints
  • Breaking down a complex problem into modular, scalable components

What’s Next for NutriLink

  • Real-time updates using WebSockets for instant synchronization
  • Route optimization for NGOs to pick up multiple donations efficiently
  • Mobile app for wider accessibility
  • Smart notifications for urgent food pickups
  • Enhanced impact analytics (meals served, CO₂ saved, etc.)
  • More advanced AI models for prediction, matching, and decision-making
  • Scaling the platform to support larger regions and organizations
Share this project:

Updates