Generative AI Customer Support Chatbot

This project implements a production-ready customer support chatbot combining fine-tuned LLaMA models with RAG architecture, deployed on AWS SageMaker with a FastAPI backend.

Features

• Context-aware responses using fine-tuned LLaMA-3
• Retrieval-Augmented Generation (RAG)
• Multi-turn conversation handling
• Responsible AI guardrails
• SageMaker and Lambda deployment
• API Gateway integration

Project Structure

• src/ : Core application code and modules
• infra/ : Infrastructure as Code (CDK)
• docker/ : Dockerfile for SageMaker deployment
• scripts/ : Helper scripts

License

MIT © January 2025 Fereydoon Boroojerdi

Installation

pip install -r requirements.txt

Usage

Run the FastAPI app:

uvicorn src.appmain:app --host 0.0.0.0 --port 8000

Running Tests

pytest tests/

Contributing

Please read CONTRIBUTING.md for details on our code of conduct and the process for submitting pull requests.

Docker Compose

To run the app locally with Docker:

docker-compose up --build

Environment Variables

Create a .env file or use sample.env as a template.

CI/CD

GitHub Actions will automatically run tests and lint checks on each pull request.

📊 Architecture Diagram