GoonSquadAI

This BFF-architected, AI-powered application takes advantage of the Strava API to acquire running data on authenticated athletes and present it in valuable ways to the users.

Pages

Main Page

This page will be comprised of...

1. A card component for each athlete that can be clicked on to go to specific stats and data visualizations for the given athlete.
2. A menu linking the following...
   1. A 'Basic Stats' page.
   2. A 'Database' page.
   3. A 'Chat' page.

Athlete Data

This page will hold specific data to the athlete selected on the main page. A user will only be directed to this page if they click on a card component from the main page.

Basic Stats

This page will hold basic week recap stats, pulled from the strava_api.activities DB table.

Database

This page will have a filterable table to hold all runs stored in the strava_api.activities DB table. The user will be able to filter by all columns.

Chat

In a traditional chatbot interface, the user will be able to ask questions about their training and receive intelligent, GenAI-driven answers. This feature will largely by driven by a TAG, or Table Augmented Generation, mechanism.

App startups

Follow these steps to get everything up and running:

1. Ensure you're using the virtual environment: .venv\Scripts\Activate.ps1.
2. In one terminal, navigate to the react directory and run npm start.
3. In another terminal, navigate to the Node server, backend.js, and run npm start.
4. In a third terminal, navigate to the FastAPI backend server, app.py, and run one of two commands:
   • Reload on code changes: uvicorn app:app --reload
   • No reload: TBD

GENERAL APP FLOW

AUTHENTICATION FLOW

1. A user authenticates with the app by clicking the 'Authenticate' button on the FE.
2. The request is routed from the FE to the BE NodeJS server, server.js.
3. The NodeJS server receives the request and routes it to the BE FastAPI server.
4. The BE FastAPI endpoint receives the request and adds the newly authenticated user to the strava_api.athletes DB table.
5. Going forward, the new user's activities will be acquired on a scheduled frequency, going into the strava_api.activities DB table.