Docaui: RAG-Based Document Assistant

# Docaui - Retrieval-Augmented Generation (RAG) Document Assistant

Docaui is an AI-powered document assistant built with the MERN stack that enables fast, contextual summarization, rewriting, and semantic search of large documents, including text, tables, and images.

## Overview

Docaui leverages Retrieval-Augmented Generation (RAG) techniques to break down complex documents and provide precise answers or summaries based on user queries. It uses FAISS for vector similarity search, ensuring sub-second retrieval speeds across large datasets. The app supports various document formats and secures user data with JWT and OAuth2.

## Key Features

- **Contextual Document Summarization:** Generate concise, relevant summaries of large documents.  
- **Semantic Search:** Perform fast, accurate natural language queries over documents with vector embeddings.  
- **Document Rewriting:** Request rewritten versions of document sections to aid understanding or repurposing.  
- **Supports Rich Content:** Works with PDFs and documents containing tables and images.  
- **Secure Authentication:** Robust backend security with JWT and OAuth2 protocols.  
- **User-Friendly Interface:** Built with React and Tailwind CSS for an intuitive experience.

## Technologies Used

- Frontend: React, Tailwind CSS  
- Backend: Node.js, Express, MongoDB  
- Vector Search: FAISS  
- Authentication: JWT, OAuth2  
- Deployment: Docker, AWS  

## Getting Started

1. Clone the repository.  
2. Install dependencies via `npm install`.  
3. Set up MongoDB and FAISS vector search service.  
4. Configure authentication secrets for JWT and OAuth2.  
5. Run the backend and frontend servers.  
6. Upload documents and start querying!

## How to Use

- **Upload Documents:** Add PDFs or text documents to the system.  
- **Query:** Ask natural language questions related to document content.  
- **Get Summaries:** Receive context-aware summaries to quickly understand sections.  
- **Rewrite:** Generate rewritten content to simplify or tailor document text.

---