Multi-AI Agentic Application Backend

This backend powers a multi-agent AI application using FastAPI, LangChain, and LangGraph. It provides robust session management, user authentication, and agent orchestration for advanced AI workflows.

Features

• FastAPI-based RESTful API: High-performance and easy-to-use API framework.
• User Authentication: Secure signup and login functionality using JWT.
• Session Management: Persistent session handling for multi-agent interactions.
• Multi-Agent Orchestration: Designed for integration with LangChain and LangGraph.
• MongoDB Integration: Asynchronous database operations using Motor.
• Structured Logging: Application-wide logging with configurable levels.
• Chat & Conversations: Chat with conversation history tracking.

Tech Stack

• Framework: FastAPI
• AI Frameworks: LangChain, LangGraph
• Database: MongoDB with Motor
• Authentication: JWT (JSON Web Tokens)
• Environment Management: python-dotenv
• LLM Providers: Google Gemini, Groq

Project Structure

backend/
├── app/
│   ├── api/            # API Route handlers (chat, sessions, users)
│   ├── auth/           # Authentication & Security logic (JWT, hashing)
│   ├── configs/        # App configurations
│   ├── db/             # Database connection & setup
│   ├── graphs/         # LangGraph workflows for AI agents
│   ├── persistance/    # Repositories for DB interactions
│   ├── schemas/        # Pydantic models for validation
│   ├── services/       # Business logic (chat, LLM, web search)
│   ├── utils/          # Helper functions (logger, error handlers)
│   └── main.py         # Entry point of the application
├── .env.example        # Example environment variables
├── requirements.txt    # Python dependencies

Getting Started

Prerequisites

• Python 3.10+
• MongoDB instance (Local or MongoDB Atlas)
• API Keys (at least one LLM provider required):
  • Google Gemini API Key (Recommended for production)
  • Groq API Key (Recommended for fastest inference)
  • Tavily API Key (Optional, for web search)

Installation

1. Navigate to the backend directory:

   cd backend

2. Create and activate a virtual environment:

   python -m venv .venv
   source .venv/bin/activate  # On Windows: .venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

Configuration

Create a .env file in the backend directory and add your configurations (refer to .env.example):

# Database Configuration
MONGODB_URL=mongodb://localhost:27017
MONGODB_DB_NAME=aiagent

# Authentication
SECRET_KEY=your_super_secret_key_change_this
ALGORITHM=HS256
ACCESS_TOKEN_EXPIRE_MINUTES=60

# LLM Provider API Keys (at least one required)
GOOGLE_API_KEY=your-google-api-key-here
GROQ_API_KEY=your-groq-api-key-here

# Web Search API Key
TAVILY_API_KEY=your-tavily-api-key-here

Required Environment Variables:

• MONGODB_URL - MongoDB connection string
• MONGODB_DB_NAME - Database name
• SECRET_KEY - Secret key for JWT token generation (use a strong random string)
• At least one LLM provider key (GOOGLE_API_KEY or GROQ_API_KEY)
• TAVILY_API_KEY - For web search functionality

Running the Application

Start the development server:

python app/main.py

The API will be available at http://localhost:8000. You can access the Interactive API docs (Swagger UI) at http://localhost:8000/docs.

API Endpoints

Health & Info

Method	Endpoint	Description	Auth Required
GET	/	Root endpoint with API information	❌
GET	/agent/health	Health check endpoint	❌

User Authentication

Method	Endpoint	Description	Auth Required
POST	/agent/users/signup	Create a new user account	❌
POST	/agent/users/login	Authenticate user and get JWT token	❌

Session Management

Method	Endpoint	Description	Auth Required
POST	/agent/sessions/create	Create a new agent session	✅

Chat & Conversations

Method	Endpoint	Description	Auth Required
POST	/agent/chat/messages	Send a message to AI agent and get response	✅
GET	/agent/chat/conversations/{session_id}/messages	Get conversation history for a session	✅