🌿 GreenLens - Enterprise B2B Sustainability Platform

Smart & Digital Innovation | Sustainability Hackathon 2026

Track your carbon. Transform your organization.

GreenLens is a production-grade, multi-tenant B2B SaaS platform that empowers modern enterprises, campuses, and organizations to track, manage, and reduce their collective carbon footprint using AI.

🌐 Live Production Application

• Frontend (Vercel): https://green-lens-tau.vercel.app
• Backend (Render): https://greenlens-backend-n3ws.onrender.com

---

🚀 Key Enterprise Features

🏢 B2B Multi-Tenancy Architecture

• True multi-tenant data isolation using a core Organization structure.
• Role-Based Access Control (RBAC): Users are assigned either admin or employee roles.
• Corporate Admin Dashboard: A dedicated, protected route for administrators to view aggregated company-wide CO₂ savings.

🛡️ Enterprise-Grade Security & Fraud Prevention

• pHash (Perceptual Hashing) Fraud Detection: Implemented ImageHash with a 30-day sliding window to prevent users from uploading visually similar receipts to cheat the leaderboard.
• DDoS & Cost Protection: Strict API rate-limiting via slowapi on expensive LLM routes (5 req/min) to prevent abuse of the Gemini Vision API.
• OOM Protection: Hard limits on UploadFile sizes (Max 5MB) to prevent memory exhaustion attacks.
• Secure Auth: JWT-based authentication with bcrypt password hashing and enforced strong password policies.

📱 Progressive Web App (PWA)

• Full PWA support allowing employees to install GreenLens directly to their iOS or Android home screen for a seamless, native-app-like receipt scanning experience.

📊 Compliance & Reporting

• 1-Click CSV Export: Administrators can instantly download a formatted CSV report of their entire organization's carbon offset activities for sustainability compliance and CSR reporting.
• Dynamic Data Visualization: High-performance charting using Recharts to visualize offset trends over time.

🤖 Gemini AI Integration

• Vision Scanning: Uses Google Gemini 1.5 Flash to automatically extract carbon footprint data from photos of food or transport receipts.
• AI Coach: Generates contextual, personalized nudges encouraging employees to make greener choices.

---

🛠️ Tech Stack

Backend

Technology	Purpose
FastAPI	High-performance async web framework
SQLAlchemy	ORM for database interactions
PostgreSQL	Relational Database (Neon)
SlowAPI	Endpoint rate limiting
ImageHash	Perceptual hashing for fraud detection
Alembic	Database migrations

Frontend

Technology	Purpose
React + Vite	Lightning-fast UI framework
Tailwind CSS	Utility-first styling
Recharts	Corporate dashboard data visualization
Vite PWA	Progressive Web App integration
Axios	Secure API communication

---

📂 Project Structure

GreenLens/
├── backend/                      
│   ├── main.py                  # Core API, Auth, Rate Limiting & Admin Routes
│   ├── models.py                # Organizations, Users, Activities, Stats
│   ├── vision.py                # pHash Fraud Detection + Gemini API
│   ├── alembic/                 # DB Migrations for Multi-Tenancy
│   └── requirements.txt          
│
├── frontend/                     
│   ├── src/
│   │   ├── pages/
│   │   │   ├── AdminDashboardPage.jsx # Protected corporate analytics
│   │   │   ├── DashboardPage.jsx      # Employee personal view
│   │   │   └── AuthPages.jsx          # Org creation & signup flow
│   │   └── api.js               # Centralized Axios interceptors
│   ├── vite.config.js           # PWA configuration
│   └── package.json             

---

⚡ Local Setup

1. Clone the repository

git clone <your-repo-url>
cd GreenLens

2. Backend Setup

cd backend
pip install -r requirements.txt
# Add .env with DATABASE_URL, SECRET_KEY, GEMINI_API_KEY
alembic upgrade head # Run migrations
python main.py

3. Frontend Setup

cd frontend
npm install
# Add .env.local with VITE_API_URL=http://localhost:8000
npm run dev

---

🤝 Team

GreenLens Development Team

• Smart & Digital Innovation Category
• Building scalable SaaS for a sustainable future.

---

Built for the Parul University Sustainability Hackathon 2026