Fantasy Football Chatbot

This repo contains a full-stack application covering integration with the Sleeper API, creation of a Streamlit UI, and deployment via AWS CDK.

For the full details, check out the article in Towards Data Science!

Demo Video

Prerequisites:

• cp .env.template .env
• Add relevant keys to .env

Local Deployment

Dependencies:

• Docker
• Python 3.11
• npm
• AWS CDK

Steps:

• cd fantasy_chatbot
• python3.11 -m venv venv (highly recommend using Python 3.11 to avoid unexpected issues)
• Activate the venv (e.g. . venv/bin/activate)
• pip install -r requirements.txt

There are 2 Dockerfiles in this directory, one for the Streamlit app and one for the LangGraph API server. The LangGraph one was generated via langgraph dockerfile api.Dockerfile. We needed to generate this to create a docker asset for the AWS CDK deployment.

• Build LangGraph image with docker build . -t fantasy-chatbot -f api.Dockerfile
• docker compose up to run the containers for Redis, Postgres, and the API (see docker-compose.yml)
• In another terminal, run the streamlit UI with streamlit run app.py

The app should now be available locally at http://localhost:8501

Deploy to AWS

• cd deploy
• npm i
• Deploy the stack with npm run cdk deploy
• The URL to the Load Balancer will be available in the Stack Outputs.