https://github.com/saad2134/fairlens
A full-stack SaaS-grade web app where teams upload a dataset + trained model (or plug in an API endpoint) and receive a comprehensive, interactive fairness audit report. Built to be genuinely enterprise-usable, not a toy.
https://github.com/saad2134/fairlens
ai-fairness bias-detection cloudrun enterprise ethical-ai fairness-auditing fairness-metrics firebase gemini googlecloudplatform machine-learning mlops responsible-ai
Last synced: 21 days ago
JSON representation
A full-stack SaaS-grade web app where teams upload a dataset + trained model (or plug in an API endpoint) and receive a comprehensive, interactive fairness audit report. Built to be genuinely enterprise-usable, not a toy.
- Host: GitHub
- URL: https://github.com/saad2134/fairlens
- Owner: saad2134
- Created: 2026-04-27T16:41:23.000Z (2 months ago)
- Default Branch: main
- Last Pushed: 2026-05-16T19:18:40.000Z (about 2 months ago)
- Last Synced: 2026-05-16T21:28:52.950Z (about 2 months ago)
- Topics: ai-fairness, bias-detection, cloudrun, enterprise, ethical-ai, fairness-auditing, fairness-metrics, firebase, gemini, googlecloudplatform, machine-learning, mlops, responsible-ai
- Language: TypeScript
- Homepage: https://fair-lensai.vercel.app
- Size: 1.04 MB
- Stars: 1
- Watchers: 0
- Forks: 0
- Open Issues: 6
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
⚖️ FairLens – Enterprise-grade AI fairness auditing platform
>
🚨 "FairLens is a full-stack SaaS platform that enables teams to upload datasets and trained models, then receive comprehensive, interactive fairness audit reports. Built for production use with model-format agnosticism, audit versioning, and enterprise compliance features."
## ✨ Key Features
| Feature | Description |
|---------|-------------|
| 🔍 **Bias Heatmap** | D3-rendered visualization of feature × protected attribute combinations with severity scoring |
| 🧪 **What-if Simulator** | Interactive prediction playground with SHAP explanations |
| 📊 **Comprehensive Metrics** | Demographic Parity, Equal Opportunity, Disparate Impact, Equalized Odds |
| 🔌 **Model Agnostic** | Support for Pickle, ONNX, and external API endpoints |
| 📜 **Audit Versioning** | Complete audit trail with dataset/model hashes for compliance |
| 📈 **Drift Detection** | Track fairness metrics across audit history |
## 🚀 Quick Start (Local)
```bash
# Clone the repository
git clone https://github.com/fairlens/fairlens.git
cd fairlens
# Start with Docker Compose (recommended)
docker-compose up -d
# Or run locally
# Backend
cd backend && pip install -r requirements.txt
uvicorn app.main:app --reload
# Frontend
cd frontend && npm install
npm run dev
```
## 🚀 Quick Deploy (100% Cloud - No Local Setup)
```bash
# 1. Install Firebase CLI
npm install -g firebase-tools
# 2. Login
firebase login
# 3. Create project at https://console.firebase.google.com
# Enable: Authentication, Firestore, Storage, Hosting
# 4. Initialize
firebase init
# Select: Hosting, Functions, Firestore, Storage
# 5. Build frontend
cd frontend && npm install && npm run build && cd ..
# 6. Deploy (uses free tier!)
firebase deploy
```
**Done!** Your app is live at `https://your-project.web.app`
See [DEPLOY.md](DEPLOY.md) for detailed instructions.
Access the UI at `http://localhost:5173`
## 📖 Documentation
- [Getting Started](https://docs.fairlens.ai/getting-started)
- [API Reference](https://docs.fairlens.ai/api-reference)
- [Fairness Metrics Guide](https://docs.fairlens.ai/metrics)
- [Deployment Guide](https://docs.fairlens.ai/deployment)
## 🏗️ Architecture
```
┌─────────────────────────────────────────────────────────────────┐
│ Frontend │
│ React 18 + TypeScript + Nextjs + TailwindCSS + Recharts + D3 │
└────────────────────────────┬────────────────────────────────────┘
│ REST API + WebSocket
┌────────────────────────────▼────────────────────────────────────┐
│ Backend │
│ FastAPI + SQLAlchemy + Celery + Redis │
├─────────────────────────────────────────────────────────────────┤
│ Fairness Engine: fairlearn | AIF360 | SHAP | ONNX Runtime │
├─────────────────────────────────────────────────────────────────┤
│ Storage: PostgreSQL + S3/MinIO │
└─────────────────────────────────────────────────────────────────┘
```
## 📦 Available Packages
| Package | Description |
|---------|-------------|
| `@fairlens/ui` | React component library |
| `@fairlens/sdk` | JavaScript SDK for API integration |
| `fairlens-core` | Python fairness computation engine |
| `fairlens-cli` | Command-line interface |
## 🔧 Tech Stack
### Frontend
- React 18 + TypeScript + Nextjs
- TailwindCSS
- Recharts (metrics visualization)
- D3.js (bias heatmap)
- Zustand (state management)
- React Query (async data)
### Backend
- Python 3.11+
- FastAPI
- SQLAlchemy + PostgreSQL
- Celery + Redis
- fairlearn, AIF360, SHAP, ONNX Runtime
### Infrastructure
- Docker & Docker Compose
- Nginx
- AWS S3 / MinIO
- GitHub Actions
## ☁️ Cloud Deploy (100% Free Tier)
For Google Solution Challenge 2026, FairLens can be deployed entirely on Google Cloud:
| Service | Free Tier | Purpose |
|---------|----------|---------|
| Firebase Hosting | 1 GB | Frontend |
| Firestore | 20K reads/writes | Database |
| Firebase Storage | 5 GB | File uploads |
| Cloud Functions | 2M invocations | API Backend |
| Gemini (Free) | 15 RPM, 1M tokens | AI Insights |
**Total Cost: $0**
### Cloud Architecture
```
Users → Firebase Hosting (Next.js)
↓
Cloud Functions (Python)
↓
Firestore + Storage
↓
Gemini API (AI Insights)
```
### Quick Deploy
See [DEPLOY.md](DEPLOY.md) for one-command deploy!
## 🤝 Contributing
We welcome contributions! Please see our [Contributing Guide](CONTRIBUTING.md) and [Code of Conduct](CODE_OF_CONDUCT.md).
```bash
# Run development environment
docker-compose -f docker-compose.dev.yml up
# Run tests
npm run test # Frontend
pytest # Backend
# Format code
npm run format # Frontend
black . # Backend
```
## 🔗 Links
- **[Live Demo](https://fairlens.ai/demo)** - Try it out
- **[Documentation](https://docs.fairlens.ai)** - Full docs
- **[Discord](https://discord.gg/fairlens)** - Community
- **[Blog](https://fairlens.ai/blog)** - Updates & guides
- **[Twitter](https://twitter.com/fairlensai)** - Follow us
## 👥 Our Google Solution Challenge 2026 Hackathon Team (DevBandits)
#
Team Member
Role
GitHub Profile
1
Abdur Rahman Qasim
🎯 Team Lead
🔗 FareedAhmedOwais
2
Fareed Ahmed Owais
🖼️ Frontend Developer
🔗 Abdur-rahman-01
3
Mohammed Saad Uddin
🚀 Full-stack + AI/ML Developer
🔗 saad2134
4
Mohammed Abdul Rahman
🎨 UI/UX Designer
🔗 saad2134
## 📊 **Repo Stats**
## ⭐ Star History
---
## ✍️ Endnote
Developed with 💖 for the Google Solution Challenge 2026 Hackathon (powered by Hack2Skill), with heartfelt thanks for the opportunity to build and innovate.
---
## 🏷 Tags
#WebApp #AI #FairnessAuditing #MachineLearning #NextJS #FastAPI #Python #TypeScript #Demo #React #TailwindCSS #BiasDetection #FairnessMetrics #EthicalAI #ResponsibleAI #GoogleCloud #Firebase #Gemini #CloudRun #AIFairness #DataScience #MLOps #OpenSource #SaaS #Enterprise #WebDevelopment