AI Chatbot Backend (FastAPI + MongoDB)

Overview

This project implements a backend service for an AI-powered chatbot system. It allows users to authenticate, create chat sessions, interact with an AI assistant, and store conversation history.

The backend is built using FastAPI and MongoDB, following a modular and scalable architecture.

Key Features

• User authentication using JWT
• Chat session management
• AI-powered chat responses
• Persistent message storage
• Modular and scalable backend design
• Interactive API documentation with Swagger

---

Tech Stack

• Backend Framework: FastAPI
• Database: MongoDB
• Async Driver: Motor
• Authentication: JWT (JSON Web Tokens)
• Password Hashing: Passlib (bcrypt)
• Validation: Pydantic
• AI Integration: Google Gemini API (optional)
• Documentation: Swagger / OpenAPI

---

Architecture

Client (Swagger / Frontend)
│
▼
API Routes (FastAPI)
│
▼
Service Layer (Business Logic)
│
▼
Database Layer (MongoDB)

Layers

• API Layer: Handles HTTP requests and responses
• Service Layer: Contains business logic
• Database Layer: Manages data persistence
• Schemas: Validate request and response data
• Core: Handles configuration and security
• Middleware: Logging and error handling

---

Project Structure

app/
├── main.py
├── core/
│   ├── config.py
│   ├── security.py
│
├── api/
│   └── routes/
│       ├── auth.py
│       ├── chat.py
│       └── sessions.py
│
├── models/
│   ├── user.py
│   ├── session.py
│   └── message.py
│
├── schemas/
│   ├── user.py
│   └── chat.py
│
├── services/
│   ├── auth_service.py
│   ├── chat_service.py
│   └── llm_service.py
│
├── db/
│   └── mongodb.py
│
├── middleware/
│   ├── logging.py
│   └── error_handler.py

---

File-Level Explanation

main.py

Initializes FastAPI application, registers routes, and loads middleware.

core/config.py

Manages environment variables such as database connection, JWT secrets, and API keys.

core/security.py

Handles JWT authentication and extracts user information from tokens.

api/routes/auth.py

Provides authentication endpoints:

• POST /auth/signup
• POST /auth/login

api/routes/sessions.py

Handles chat session creation:

• POST /sessions

api/routes/chat.py

Handles chatbot interaction:

• POST /chat/{session_id}
• GET /sessions/{session_id}

models/

Defines data structures for users, sessions, and messages.

schemas/

Defines request and response validation using Pydantic.

services/

Contains business logic:

• auth_service.py → authentication logic
• chat_service.py → chat handling
• llm_service.py → AI response generation

db/mongodb.py

Initializes MongoDB connection and collections.

middleware/

Handles logging and error management.

---

API Documentation

Swagger UI is available at:

http://127.0.0.1:8000/docs

It provides:

• Interactive API testing
• Request/response schemas
• Authorization support

---

API Endpoints

Authentication

Signup

POST /auth/signup

Request:

{
  "email": "user@example.com",
  "password": "password123"
}

Login

POST /auth/login

Response:

{
  "access_token": "JWT_TOKEN"
}

---

Sessions

Create Session

POST /sessions

Response:

{
  "session_id": "session_id"
}

---

Chat

Send Message

POST /chat/{session_id}

Request:

{
  "message": "Hello AI"
}

Response:

{
  "reply": "AI response"
}

---

Get Chat History

GET /sessions/{session_id}

Response:

{
  "session_id": "...",
  "messages": [...]
}

---

Database Schema

Users Collection

{
  "_id": ObjectId,
  "email": "string",
  "password_hash": "string"
}

Sessions Collection

{
  "_id": ObjectId,
  "user_id": "string",
  "created_at": "timestamp"
}

Messages Collection

{
  "_id": ObjectId,
  "session_id": "string",
  "role": "user | assistant",
  "content": "string",
  "timestamp": "timestamp"
}

---

Setup Instructions

1. Clone Repository

git clone <repository-url>
cd ai-chatbot-backend

2. Create Virtual Environment

python -m venv venv

Activate (Windows):

venv\Scripts\activate

---

3. Install Dependencies

pip install -r requirements.txt

---

4. Environment Variables

Create a .env file:

MONGO_URL=mongodb://localhost:27017
DATABASE_NAME=ai_chatbot_db

JWT_SECRET=your_secure_secret
JWT_ALGORITHM=HS256

LLM_PROVIDER=mock
GEMINI_API_KEY=

---

5. Run Server

uvicorn app.main:app --reload

Visit:

http://127.0.0.1:8000/docs

---

Assumptions

• Each user can create multiple chat sessions
• Each session represents one conversation
• Messages are stored separately and linked via session_id
• JWT is used for authentication
• Passwords are hashed using bcrypt
• MongoDB is used for flexible schema handling
• AI provider can be switched between mock and real

---

Advanced Feature

Context-Aware Chat Memory

Example:

User: What is Python?
AI: Python is a programming language.

User: Who created it?

With context:

AI: Python was created by Guido van Rossum.

---

Future Improvements

• Streaming responses
• Rate limiting
• Chat summarization
• Multi-user chat support
• AI moderation system

---

Conclusion

This project demonstrates a scalable and modular backend system for AI-powered chat applications. It follows best practices in API design, authentication, and database handling, making it suitable for real-world applications and further extension.