NBA betting-intelligence SaaS — ML predictions, line shopping, bankroll tracking.
In active build
XGBoost + LightGBM + scikit-learn retrained on schedule via Celery + Redis. Multi-sportsbook odds aggregation for line shopping; bankroll tracking tied to live game state.
Overview
An NBA betting-intelligence SaaS. Backend is FastAPI + async SQLAlchemy on Postgres, with Celery workers running on Redis to ingest live game data via nba_api and to retrain prediction models (XGBoost + LightGBM) on a schedule. Surfaces line shopping across sportsbooks, bankroll + bet-history tracking, and live scores tied to open positions. React 19 + Vite frontend with TanStack Query and Recharts. Stripe-billed.
Highlights
- 01ML prediction stack — XGBoost + LightGBM + scikit-learn — retrained on schedule
- 02Async FastAPI + SQLAlchemy + Postgres backend; Celery + Redis for background workers
- 03Live NBA data via nba_api; multi-sportsbook odds aggregation for line shopping
- 04Bankroll + bet-history tracking with live scores tied to open positions
- 05Stripe-billed SaaS, JWT auth (python-jose + passlib)
- 06React 19 + Vite + TanStack Query + Recharts frontend
Stack
Python 3.12FastAPISQLAlchemy (async) + PostgreSQLRedis + CeleryXGBoost + LightGBM + scikit-learnnba_apiStripeReact 19 + Vite + TypeScript