README.md

---
title: RAG System with Ollama Cloud
emoji: 🤖
colorFrom: blue
colorTo: purple
sdk: docker
pinned: false
---

# RAG (Retrieval-Augmented Generation) QA System

This is a production-ready RAG system using:
- **MongoDB Atlas Vector Search** for document storage and retrieval
- **Ollama Cloud (gpt-oss:120b-cloud)** for text generation
- **Sentence Transformers** for embeddings
- **FastAPI** for the web interface

## Features
- Upload and process PDF, CSV, JSON documents
- Scrape and index web pages
- Semantic search using MongoDB Atlas
- Question answering using Ollama Cloud models

## Configuration
Set these environment variables in your Space settings:
- `OLLAMA_API_KEY`: Your Ollama Cloud API key
- `MONGODB_URI`: MongoDB Atlas connection string