Serverless AI Chatbot MVP

A scalable, serverless AI chatbot backend built with AWS SAM, featuring hexagonal architecture for maintainability and portability.

⚠️ Security Disclaimer

This code is for demonstration purposes only and lacks the authorization and security measures required for a production MVP. Before deploying to production, implement at least:

• API authentication (API keys, JWT tokens, or AWS Cognito)
• Rate limiting and throttling
• Input validation and sanitization
• Proper IAM roles with least privilege
• Monitoring and logging for security events

Architecture

• API Gateway: REST endpoint (⚠️ No authentication in demo)
• Lambda Function: Stateless chat orchestrator
• DynamoDB: Conversation history storage
• Amazon Bedrock: AI model integration (Claude 3 Haiku)

Project Structure

src/
├── app.py                    # Lambda handler (orchestrator)
├── core/
│   └── chat_service.py      # Framework-agnostic business logic
├── adapters/
│   ├── database_adapter.py  # DynamoDB operations
│   └── llm_adapter.py       # Bedrock AI integration
└── requirements.txt

Prerequisites

• AWS CLI configured with appropriate permissions
• AWS SAM CLI installed
• AWS Account with anthropic.claude-3-haiku-20240307-v1:0 model enabled in Amazon Bedrock. (If you prefer to use a different one make appropriate changes in src/adapters/llm_adapter.py)
• Python 3.11+

Deployment

1. Build the application:

   sam build

2. Deploy to AWS:

   sam deploy --guided

3. Note the API endpoint URL from the deployment output.

API Usage

Endpoint

POST /chat

Request Body

{
  "userId": "user123",
  "message": "Hello, how are you?"
}

Success Response (200)

{
  "aiResponse": "Hello! I'm doing well, thank you for asking. How can I help you today?"
}

Error Responses

• 400 Bad Request: Invalid input
• 500 Internal Server Error: System error

Testing

Using curl

curl -X POST https://your-api-id.execute-api.region.amazonaws.com/Prod/chat \
  -H "Content-Type: application/json" \
  -d '{"userId": "test-user", "message": "Hello!"}'

Using Postman

1. Create a new POST request
2. Set URL to your API endpoint + /chat
3. Set Content-Type header to application/json
4. Add request body with userId and message

Performance Features

• Warm Initialization: AWS SDK clients initialized outside handler
• Efficient Queries: DynamoDB queries limited to recent messages
• Stateless Design: No memory state between invocations

Scaling Considerations

For high-traffic scenarios, consider:

• Enabling Provisioned Concurrency in template.yaml
• Implementing DynamoDB auto-scaling
• Adding CloudWatch monitoring and alarms

Frontend

An example web interface is available in the frontend/ directory:

1. Deploy backend first (see Deployment section above)
2. Update API endpoint in frontend/script.js
3. Serve locally:

   cd frontend
   python -m http.server 8000

4. Open: http://localhost:8000

Clean Up

To remove all resources:

sam delete