AI

Green Ledger / Project GLIA

AI Nutrition Intelligence

The "Google Maps for Nutrition" - an AI-powered app that reveals invisible quality differences in groceries using supply chain data.

Tech Stack

Google Vision AI
Next.js
Supabase
React Native

The Challenge

The "Olive Oil Problem": Two bottles of olive oil, identical on the shelf, can have a 20x difference in nutritional value. Harvest date, container type, processing methods, and storage conditions all dramatically affect quality—but consumers have no way to see these invisible differences.

Green Ledger needed to build a system that could surface supply chain truth and calculate a "Nutrition per Dollar" score for every product.

Our Solution

Vision AI Scanner

Integrated Google Vision AI (Gemini 1.5 Pro) to scan product labels, extract harvest dates, certifications, and nutritional claims with OCR, enabling instant analysis.

Trust Waterfall System

Built a weighted trust algorithm: Lab Data (100%) → User Proof (90%) → Label Claim (40%) → AI Guess (20%), ensuring data reliability and transparency.

React Native Mobile App

Created a mobile-first experience with camera integration, offline sync, GPS location tracking, and beautiful data visualizations using Framer Motion.

Supabase Backend

Leveraged PostgreSQL with Row-Level Security for user data, product database, and supply chain records, with real-time sync across devices.

The Results

Computer vision AI integration with Google Gemini
Supply chain data tracking and verification
"Nutrition per Dollar" scoring algorithm
Mobile-first React Native app with offline sync

Three-Phase Roadmap

Phase 1: Consultant

Personal nutrition guide with product recommendations

Phase 2: Scanner

Camera-based product analysis and scoring

Phase 3: Platform

Community-driven supply chain verification