RAG PDF Legal Document Chat Application

Overview

This application provides an intelligent, context-aware chatbot for legal document analysis using Retrieval Augmented Generation (RAG) with Pinecone vector storage and Google Gemini AI.

Features

• 📄 Upload and process legal documents
• 🔍 Semantic search through document content
• 💬 AI-powered query answering
• 🧠 Contextual understanding of legal documents

Technologies Used

• Next.js
• TypeScript
• Pinecone Vector Database
• Google Gemini AI
• Hugging Face Inference
• Vector Embedding

Prerequisites

• Node.js (v18+)
• Pinecone Account
• Google AI Studio Account
• Hugging Face Account

Environment Setup

1. Copy .env.example to .env
2. Fill in the values with your actual credentials
3. Do not commit your .env file

PINECONE_API_KEY=your_pinecone_api_key
HF_TOKEN=your_huggingface_token
GEMINI_API_KEY=your_google_ai_studio_key

Installation

1. Clone the repository

git clone https://your-repo-url.git

2. Install dependencies

npm install

3. Run the development server

npm run dev

Key Components

Vector Storage (utils.ts)

• Handles embedding generation
• Queries Pinecone vector store
• Retrieves contextually relevant document sections

Route Handler (route.ts)

• Processes user queries
• Integrates document context
• Streams AI-generated responses

Embedding Model

Uses mixedbread-ai/mxbai-embed-large-v1 for high-quality semantic embeddings

Deployment Considerations

• Ensure vector index is pre-populated
• Configure proper environment variables
• Use serverless/edge runtime compatible deployment

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.