GrainGain

AI-powered food rescue: Connect surplus restaurant meals to hungry communities before waste happens.

Comment

Inspiration

Every day, cities waste millions of tons of edible food while NGOs and food-insecure communities struggle to access fresh meals. We were inspired by a simple question: What if surplus food could find hungry people in minutes, not days? We imagined an AI-powered system that acts as a bridge between restaurants with excess inventory and nonprofits with urgent need. GrainGain emerged from the conviction that technology can solve real-world social problems with elegance and speed.

What it does

GrainGain is an AI-powered smart city platform that enables restaurants to instantly broadcast surplus food to nearby NGO partners. Users describe food, the AI analyzes freshness and urgency, and a map-based interface connects them with vetted nonprofits ready to pick up. The app calculates logistics (distance, ETA, food safety windows) and streamlines the request workflow — turning food waste into measurable social impact in real time.

How we built it

  • Frontend: React with custom CSS animations and scroll-driven interactions for cinematic storytelling
  • Maps & Location: Leaflet integration for interactive NGO and restaurant discovery
  • AI Analysis: Featherless AI (Qwen3-4B) to analyze food descriptions and estimate safe time windows
  • Backend: Node.js server handling food analysis requests and logistics calculations
  • Deployment: Frontend on Netlify, backend on Render
  • Data: Hardcoded NGO and restaurant datasets with geolocation for MVP proof-of-concept

Challenges we ran into

  • Real-time accuracy: Balancing AI food analysis speed with accuracy — optimized for sub-second response times
  • Map complexity: Managing interactive selections, filtering, and real-time updates without performance degradation
  • Logistics math: Calculating realistic ETAs and food safety windows for accurate urgency signaling
  • Responsive design: Creating a premium, cinematic experience across mobile, tablet, and desktop
  • API latency: Coordinating AI inference with smooth UI feedback during loading states

Accomplishments that we're proud of

  • Built a fully functional MVP with end-to-end AI workflow in limited time
  • Created a cinematic, story-driven user experience that tells the impact narrative
  • Integrated live maps with intuitive NGO selection and restaurant discovery
  • Designed an AI-powered food analyzer that contextualizes urgency and safety
  • Deployed both frontend and backend to production with live demo ready
  • Implemented responsive design that feels premium across all devices

What we learned

  • AI doesn't solve logistics problems alone — context, UI clarity, and urgency signals matter equally
  • Story-driven design can make civic tech engaging and shareable
  • Simple data models (hardcoded NGO/restaurant lists) are sufficient for MVP validation
  • Fast, predictable AI inference is crucial for real-time user-facing applications
  • Animated transitions and visual hierarchy dramatically improve conversion in social impact apps

What's next for GrainGain

  • Database integration: Move from hardcoded data to real NGO and restaurant databases with live availability
  • Notification system: SMS/email alerts to NGOs when food matches their needs
  • Food safety scoring: Machine learning model trained on real food waste data to improve urgency predictions
  • Pickup tracking: Real-time GPS tracking and photo evidence of successful food rescue
  • Impact quantification: Blockchain or secure logging of meals rescued, CO₂ avoided, and community reach
  • Mobile app: Native iOS/Android for faster restaurant and NGO adoption
  • Partnership scaling: Integration with local food recovery networks and city governments

Built With

Share this project:

Updates