GreenScore is a barcode-scanning tool and search assistant that helps users make more ethical and sustainable food choices — powered by AI and built for impact.
We were inspired by the rising demand for ethical consumerism and the growing frustration around how difficult it is to make truly informed shopping decisions.
Greenwashed packaging, vague eco-labels, and hidden supply chain issues often leave well-intentioned shoppers in the dark.
GreenScore aims to bring clarity and accountability to the food industry — helping users align their purchases with their values in a simple, confident way.
GreenScore analyzes a product’s environmental footprint, labour ethics, ingredient sourcing, and health impact, and returns:
- ✅ A GreenScore (0–100)
- 📋 A concise summary of sustainability, ethics, and health factors
- 📉 A list of pros, cons, and verified facts
- 🔁 Recommended alternatives with better scores
- 💬 An AI-powered chatbot assistant that answers questions and suggests better options on the go
Users can either:
- 📷 Scan a barcode using their device camera
- 🔍 Search a product name to get instant feedback
We designed GreenScore with a modular full-stack architecture to maximise speed, scalability, and future growth:
- Frontend: React + Tailwind CSS (clean, responsive UI)
- Backend: FastAPI with prompt-engineered Gemini AI integration
- Database: MongoDB Atlas (stores users, queries, and scores)
- Authentication: Firebase Auth (secure login & access)
Product scores are generated via Gemini AI, using structured prompts that map verified data to a transparent scoring rubric.
- Finding reliable, structured data for sustainability and ethical sourcing
- Crafting AI prompts that were factual, consistent, and parsable
- Building cross-platform, browser-based barcode scanning
- Creating a polished UI under tight hackathon constraints
- Delivered a fully functional demo with real-world value
- Integrated AI for complex, explainable scoring
- Created a meaningful user experience focused on impact and ease-of-use
- How to harness LLMs like Gemini in a controlled, useful way
- That frictionless design is key to encouraging sustainable decisions
- Effective team collaboration to ship a full-stack AI app in under 48 hours
- Expanding the product database with verified and crowdsourced data
- Integrating MCP for real-time score syncing with product updates
- Developing a browser extension for instant scoring while shopping online
- 🧠 Gemini
- ⚙️ FastAPI
- 🧾 MongoDB Atlas
- 🔐 Firebase Auth
- 💻 React
- 💅 Tailwind CSS
- 🐍 Python
Made with purpose at GDG HACKS 2025 — 32 hours of caffeine, code, and climate-conscious energy 🌍💚
Contributors: Rehan Siddiqi, Yousuf Mohiuddin, Muzammil Zaman, and Adil Siddiqi