🧠 Nexus RAG — Artificial Intelligence Augmented

Nexus RAG is a minimalist, high-performance Retrieval Augmented Generation (RAG) platform. It allows you to upload PDF documents and interrogate them using semantic search and cutting-edge LLMs.

🚀 Quick Start

1. Setup Environment:

   • Clone the repository.
   • Create a virtual environment: python -m venv .venv
   • Activate it: .venv\Scripts\activate (Windows) or source .venv/bin/activate (Mac/Linux)
   • Install dependencies: pip install -r requirements.txt

2. Configure API:

   • Create a .env file based on .env.example.
   • Add your GROQ_API_KEY.

3. Run Application:

   python run.py

   Open http://localhost:5000 in your browser.

📁 Project Structure

.
├── app/                    # Main application package
│   ├── api/                # API blueprints and routes
│   │   └── routes.py
│   ├── core/               # Core business logic (RAG pipeline)
│   │   └── rag_engine.py
│   ├── static/             # Frontend assets (CSS, JS)
│   ├── templates/          # HTML templates
│   └── __init__.py         # App factory
├── uploads/                # Temporary directory for processed PDFs
├── run.py                  # Entry point
├── .env                    # Configuration
└── requirements.txt        # Dependencies

🛠️ Tech Stack

• Backend: Flask, LangChain, FAISS
• Embeddings: Sentence-Transformers (all-MiniLM-L6-v2)
• LLM: Llama 3 (via Groq)
• Frontend: Vanilla JS, CSS (Quantum Dark Theme)

☁️ Deployment (Render)

1. New Web Service: Connect your GitHub repo to Render.
2. Environment Variables:
   • GROQ_API_KEY: Your key from Groq Console.
   • GROQ_MODEL: llama-3.1-8b-instant (optional).
   • PYTHON_VERSION: 3.11.0 (recommended).
3. Build Command: pip install -r requirements.txt
4. Start Command: gunicorn run:app

🔺 Deployment (Vercel)

1. New Project: Import your GitHub repo into Vercel.
2. Framework Preset: Choose "Other" (it will auto-detect vercel.json).
3. Environment Variables:
   • GROQ_API_KEY: Your key.
   • VERCEL: 1 (required for temp directory switching).
4. Deploy: Vercel will build the serverless function and serve your app.

🌐 Split Hosting (Recommended for Speed)

If you want the fastest UI response, host the Frontend as a static site and the Backend on Render.

1. Frontend: Upload the frontend/ folder to Vercel or Netlify.
2. Configuration: In frontend/js/app.js, update API_BASE_URL to your Render URL.
3. Backend: Keep the main repository running on Render.

⚖️ License

MIT