📄 PDF Question Answering System (RAG-based)

A PDF Question Answering web application that allows users to upload any PDF document and ask natural language questions.
The system uses Semantic Search + Embeddings + FAISS + LLM to generate accurate, context-aware answers.

This project demonstrates Retrieval-Augmented Generation (RAG) and is suitable for real-world, production-style use cases.

---

🚀 Features

• Upload any text-based PDF
• Ask custom questions (not limited to predefined ones)
• Semantic search using Sentence Transformers
• Fast retrieval with FAISS vector database
• AI-generated answers in 2–3 bullet points
• Clean Streamlit web interface
• Works with multiple PDFs (one at a time)

---

🧠 Tech Stack

• Python
• Streamlit – Web app UI
• PyPDF – PDF text extraction
• Sentence-Transformers – Text embeddings
• FAISS – Vector similarity search
• OpenAI API – Answer generation
• dotenv – Secure API key management

🚀 Live Demo

🔗 Streamlit App:
https://genai-pdf.streamlit.app/