Swish Report is a full-stack web app for basketball player analysis across High School, College, and NBA levels. It aggregates data, generates AI scouting reports, and lets users build/comparing lineups—wrapped in a modern, scalable stack.
- Frontend: React + Next.js (TypeScript)
- Backend: Python + FastAPI
- Playwright (web scraping)
- OpenAI + Gemini (LLM integrations)
- OpenCV (generating highlight reels and basketball player scouting videos)
- Database: MySQL (relational modeling)
- Infrastructure: Docker (containerization)
- AI scouting reports for players at HS/College/NBA (OpenAI & Gemini powered)
- AI generated highlight reels for players at every level of basketball (OpenCV & Gemini powered)
- Player pages: Scouting reports, player related content and highlight reels, and forums to discuss player skillset and potential
- Auth: Email/password + Google OAuth signup and login
- Lineup builder game with interactive team composition and lineup analysis
- Community: Compare lineups, post takes/hot-takes, discuss player scouting analysis
- Expand scraping coverage for college players and international players
- Deeper personalization and richer community features (follows, votes, badges)
- More interactive games (salary cap drafts, scenario simulators, p)
- Deploy backend and database for the application on a cloud service like AWS, GCP, or Azure
- Next.js (app or pages router) serves the UI + API proxy where needed
- FastAPI exposes REST endpoints for player search, reports, lineups, and community
- MySQL stores player master data and metadata, evaluations, sources, users, and community content
- Playwright scrapers feed normalized data into MySQL via ETL jobs
- LLM layer (OpenAI/Gemini) summarizes scouting text and produces structured insights
- Docker standardizes local/dev/prod environments