RAG Document Search

A lightweight Retrieval-Augmented Generation (RAG) application that enables semantic document search with LLM-powered answers. The system indexes documents into a vector store, retrieves relevant context using FAISS, and generates accurate responses via a language model — all wrapped in a Streamlit UI.

🚀 Features

• 📚 Document ingestion and chunking
• 🔎 Semantic search using vector embeddings
• 🧠 Retrieval-Augmented Generation (RAG) pipeline
• ⚡ Fast similarity search with FAISS
• 🎛️ Interactive frontend built with Streamlit
• 🧩 Modular and easy-to-extend architecture

🏗️ Architecture Overview

1. Document Processing
   • Load and split documents into chunks
2. Embeddings & Vector Store
   • Generate embeddings
   • Store vectors using FAISS
3. Retrieval
   • Retrieve top-k relevant chunks for a query
4. Generation
   • Pass retrieved context to an LLM
   • Generate grounded answers
5. Frontend
   • Streamlit UI for querying documents

🖥️ Tech Stack

• Python
• Streamlit – Frontend UI
• FAISS – Vector store
• LangChain / LangGraph
• LLM Provider (OpenAI)
• Embedding Models (OpenAI)

📦 Installation

git clone https://github.com/saadtariq-ds/rag-document-search.git
cd rag-document-search

Create a virtual environment (recommended)

uv init
uv venv
.venv\Scripts\activate

Install dependencies

uv add -r requirements.txt

🔑 Environment Variables

Create a .env file and add your API keys if required:

OPENAI_API_KEY=your_api_key_here

(Adjust based on the LLM provider you are using.)

▶️ Run the Application

streamlit run app.py

Then open your browser at:

http://localhost:8501