🤖 Agentic RAG - PDF Q&A

AI-Powered PDF Question Answering System using CrewAI and Groq

Upload any PDF and ask questions - get complete, AI-generated answers powered by advanced language models!

🌟 Features

• 📄 PDF Upload & Search - Upload any PDF document and ask questions
• 🤖 AI-Powered Answers - Complete, coherent responses using state-of-the-art LLMs
• 🎯 Smart Search - Searches PDF first, then web if needed (optional)
• 🔑 User-Provided API Keys - Secure, no hardcoded secrets
• 🎨 Model Selection - Choose from multiple Groq models
• 💰 100% FREE - Uses FREE Groq API (no OpenAI costs!)
• 🚀 Fast & Reliable - Powered by Groq's lightning-fast inference
• 🔒 Privacy-Focused - Your API keys and documents are never stored

🎯 Live Demo

Deploy your own!

🚀 Quick Start

1. Get FREE API Key

Get your FREE Groq API key (no credit card required):

• Go to: console.groq.com
• Sign up for FREE
• Get your API key

2. Run Locally

# Clone the repository
git clone https://github.com/YOUR_USERNAME/agentic-rag-pdf-qa.git
cd agentic-rag-pdf-qa

# Install dependencies
pip install -r requirements.txt

# Run the app
streamlit run app_final.py

3. Use the App

1. Enter your Groq API key in the sidebar
2. Select a model (Llama 3.3 70B recommended)
3. Upload a PDF document
4. Ask questions and get AI-powered answers!

📦 Installation

Requirements

• Python 3.10 or higher
• pip

Install Dependencies

pip install -r requirements.txt

Dependencies Include:

• streamlit - Web interface
• crewai - AI agent framework
• markitdown - PDF text extraction
• chonkie - Semantic chunking
• qdrant-client - Vector database
• fastembed - Fast embeddings
• python-dotenv - Environment variables

🌐 Deploy to Streamlit Cloud

1. Push to GitHub

   git init
   git add .
   git commit -m "Initial commit"
   git push origin main

2. Deploy on Streamlit Cloud

   • Go to share.streamlit.io
   • Connect your GitHub repository
   • Select app_final.py as the main file
   • Deploy!

3. No Secrets Needed!

   • Users will enter their own API keys in the UI
   • No need to configure secrets in Streamlit Cloud

🛠️ Tech Stack

• Frontend: Streamlit
• LLM: Groq API (Llama 3.3 70B, Mixtral, Gemma)
• PDF Processing: MarkItDown
• Vector Search: Qdrant + Chonkie
• AI Agents: CrewAI
• Web Search: FireCrawl (optional)

📖 How It Works

1. PDF Upload: User uploads a PDF document
2. Indexing: PDF is converted to text and chunked semantically
3. Vector Storage: Chunks are stored in Qdrant vector database
4. Question: User asks a question
5. Retrieval: AI agent searches PDF for relevant information
6. Synthesis: AI synthesizes a complete answer from retrieved chunks
7. Response: User gets a coherent, AI-generated answer

🎨 Available Models

• Llama 3.3 70B (Recommended) - Best quality, comprehensive answers
• Llama 3.1 8B - Faster responses, good quality
• Mixtral 8x7B - Excellent reasoning capabilities
• Gemma 2 9B - Balanced performance

💡 Why This App?

✅ No OpenAI costs - Uses FREE Groq API
✅ Complete answers - Not just chunks, full AI synthesis
✅ Privacy-focused - API keys never stored
✅ Fast - Groq's inference is lightning fast
✅ Easy to deploy - One-click Streamlit Cloud deployment
✅ User-controlled - Users provide their own API keys
✅ Model choice - Select the best model for your needs

🔒 Privacy & Security

• ✅ API keys are never stored - only used during your session
• ✅ PDFs are processed locally in temporary storage
• ✅ No data is sent to external servers except Groq API
• ✅ Session data is cleared when you close the browser

📝 Project Structure

agentic-rag-pdf-qa/
├── app_final.py              # Main Streamlit application
├── requirements.txt          # Python dependencies
├── README.md                 # This file
├── DEPLOYMENT_GUIDE.md       # Detailed deployment instructions
├── .gitignore               # Git ignore rules
├── knowledge/               # Sample PDF files
│   └── dspy.pdf            # Example document
└── src/
    └── agentic_rag/
        └── tools/
            └── custom_tool.py  # Custom search tools

🤝 Contributing

Contributions welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Submit a pull request

📄 License

MIT License - Feel free to use and modify!

🙏 Acknowledgments

• CrewAI - AI agent framework
• Groq - FREE, fast LLM inference
• Streamlit - Amazing web framework
• Qdrant - Vector database
• Chonkie - Semantic chunking

⭐ Star this repo if you find it useful!

---

Built with ❤️ for the AI community