PDF-chat-RAG AI - Intelligent PDF Conversational Assistant

A production-ready RAG (Retrieval-Augmented Generation) application that enables intelligent conversations with multiple PDF documents using advanced NLP techniques and Groq's high-performance LLM inference.

🚀 Features

• Multi-Document Processing: Upload and analyze multiple PDFs simultaneously
• Intelligent Chunking: Smart text segmentation with TF-IDF vectorization for optimal retrieval
• Semantic Search: Cosine similarity-based context retrieval for precise answers
• Conversational Memory: Maintains context across multi-turn conversations
• Streaming Responses: Real-time token generation for enhanced UX
• Source Attribution: Tracks and displays which documents provided the answer

🛠️ Technical Architecture

┌─────────────┐      ┌──────────────┐     ┌─────────────┐
│  PDF Upload │────▶│ Text Extract │────▶│   Chunking  │
└─────────────┘      └──────────────┘     └──────┬──────┘
│
┌─────────────┐      ┌──────────────┐     ┌──────▼──────┐
│   Groq LLM  │◀────│ Context Build│◀────│ TF-IDF +    │
│   Response  │      │   + Prompt   │     │ Similarity  │
└─────────────┘      └──────────────┘     └─────────────┘

📦 Installation

1. Download the code

   • Click the green "Code" button above
   • Click "Download ZIP"
   • Extract the ZIP file

2. Install Python (3.8 or higher) from python.org

3. Open Command Prompt in the folder

   • Navigate to the extracted folder
   • Type cmd in the address bar and press Enter

4. Create virtual environment

   python -m venv venv
   venv\Scripts\activate
   

5. Install dependencies

   pip install -r requirements.txt
   

6. Set up API Key

   • Create a file named .env
   • Add: GROQ_API_KEY=your_api_key_here

7. Run the application

   streamlit run Main.py
   

🔑 Get API Key

1. Go to console.groq.com
2. Create free account
3. Generate API key
4. Paste in .env file

💡 How It Works

Upload PDFs: Drag and drop multiple documents
Process: Click "Process PDFs" to index content
Chat: Ask questions in natural language
Review: Check source references for transparency

🧠 Technologies Used

Streamlit: Web interface framework
Groq: Ultra-fast LLM inference (Llama 3.1, Mixtral)
PyPDF2: PDF text extraction
scikit-learn: TF-IDF vectorization and similarity search
python-dotenv: Environment variable management

🛡️ Privacy & Security

Documents processed locally in memory only
No persistent storage of uploaded files
API keys managed via environment variables

📄 License

MIT License - see LICENSE file for details

🙏 Acknowledgments

Groq for high-speed LLM inference
Streamlit for intuitive UI framework
scikit-learn for ML utilities

Built with ❤️ by Badal Patel