RAG Docs

Document intelligence platform with multi-doc upload, AI-powered Q&A, semantic search, and chat history using RAG.

🌟 Features

• Multi-document upload & management
• AI-powered Q&A across documents
• Semantic search via Pinecone vectors
• Persistent chat history
• Secure JWT authentication
• Real-time document processing

🛠 Tech Stack

Frontend: Next.js 14+, React, Redux, Tailwind CSS | Backend: Express.js, TypeScript, MongoDB | AI: LangChain, Pinecone, GOOGLE | Python: FastAPI

📁 Project Structure

├── backend/          # Express API, auth, chats, messages
├── frontend/         # Next.js app, UI, Redux state
├── genai/            # FastAPI, document processing, Q&A
└── README.md

🚀 Quick Setup

Backend:

cd backend && npm install
# .env: MONGODB_URI, JWT_SECRET, PINECONE_API_KEY, OOGLE_API_KEY, PORT=5000
npm run dev      # Runs on http://localhost:5000

Frontend:

cd frontend && npm install
# .env.local: NEXT_PUBLIC_API_URL=http://localhost:5000
npm run dev      # Runs on http://localhost:3000

GenAI:

cd genai && pip install -r requirements.txt
# .env: PINECONE_API_KEY, GOOGLE_API_KEY, MONGODB_URI
python -m uvicorn app.main:app --reload    # Runs on http://localhost:8000

🔌 Key Endpoints

Endpoint	Method	Purpose
/api/users/register	POST	Register user
/api/users/login	POST	Login user
/api/chats	GET	Get all chats
/api/chats	POST	Create chat
/api/messages	POST	Send message/question
/api/documents/upload	POST	Upload document
/api/ask-question	POST	Ask AI question

📖 Quick Guide

1. Create Account → Register at /register
2. Login → Go to /login
3. Create Chat → Click "New Chat"
4. Upload Docs → Drag & drop or select files
5. Ask Questions → Type in chat input, get AI answers

⚡ Common Issues

• MongoDB error → Check MONGODB_URI and MongoDB is running
• API connection fails → Verify NEXT_PUBLIC_API_URL and backend running
• Google error → Check GOOGLE_API_KEY is valid
• Slow responses → Check Pinecone quota and network

📚 Documentation

• Backend README
• Frontend README
• GenAI README
• Privacy Policy
• Learning Guide

🔐 Security

• Passwords hashed with bcryptjs
• JWT token authentication
• Input validation & CORS protection
• User data encrypted at rest

📝 Environment Setup

All 3 services need these keys:

• PINECONE_API_KEY (from Pinecone dashboard)
• GOOGLE_API_KEY (from GOOGLE GEMINI)
• MONGODB_URI (local or MongoDB Atlas)

See individual service READMEs for full env setup.

🤝 Contributing

1. Fork repo → Create branch → Make changes → Push → Pull request
2. Follow TypeScript best practices
3. Test changes before submitting

📄 License

MIT License

🆘 Support

• Issues: GitHub Issues
• Author: @ADCobra — bondlicensedtokillqueen@gmail.com