HR GPT

This repository demonstrates how to use LangGraph with MongoDB for building and managing AI agents and conversational applications using an agentic approach. It showcases the integration of language models, graph-based conversation management, and MongoDB for data persistence, enabling the creation of intelligent, autonomous agents in TypeScript and Express.js.

Features

• Utilizes LangGraph for managing agentic conversational flows in TypeScript
• Integrates with MongoDB Atlas for storing and retrieving conversation data
• Implements a RESTful API using Express.js for chat interactions
• Uses Mistral's AI model for generating responses
• Includes a tool for employee lookup using MongoDB Atlas vector search

Prerequisites

• Node.js and npm
• MongoDB Atlas account
• Mistral API key

Installation

1. Clone this repository:

git clone https://github.com/jayesh-patidar/hr-gpt.git
cd hr-gpt

2. Install the required dependencies:

npm install

3. Set up your environment variables:

• Copy .env.example to .env file in the root directory

• Add your API keys and MongoDB URI:

  MISTRAL_API_KEY=your_mistral_api_key_here
  MONGODB_URI=your_mongodb_atlas_uri_here
  

Seed the Database

1. Run the seed script to generate and seed the database:

npm run seed

Usage

1. Start the server:

npm run dev

2. Use the following API endpoints:

• Start a new conversation:

  curl -X POST -H "Content-Type: application/json" -d '{"message": "Your message here"}' http://localhost:3000/chat
  

• Continue an existing conversation:

  curl -X POST -H "Content-Type: application/json" -d '{"message": "Your follow-up message"}' http://localhost:3000/chat/{threadId}
  

Project Structure

• index.ts: Entry point of the application, sets up the Express server and API routes
• agent.ts: Defines the LangGraph agent, tools, and conversation flow
• seed-database.ts: Script for generating and seeding synthetic employee data into MongoDB

How it works

1. The seed script in seed-database.ts generates synthetic employee data and populates the MongoDB database.
2. The LangGraph agent is defined in agent.ts, including the conversation graph structure and tools.
3. MongoDB operations are integrated directly into the agent for storing and retrieving conversation data.
4. The Express server in index.ts provides API endpoints for starting and continuing conversations.
5. User inputs are processed through the LangGraph agent, generating appropriate responses and updating the conversation state.
6. Conversation data is persisted in MongoDB Atlas, allowing for continuity across sessions.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.