AI Automation & Consulting
A baseball simulation engine, built from scratch
ClaudeBall
The problem
Not a client problem — a demonstration of engineering range. Simulating a full baseball game correctly means modeling pitch outcomes, contact quality, fielding, and baserunning as a coherent, reproducible system, then proving the numbers it produces look like real baseball.
What was built
- Deterministic, seeded simulation engine: pitch → contact → fielding → baserunning
- Statistical calibration against real MLB ranges
- Playwright visual-regression QA discipline applied to a simulation, not a typical web app
By the numbers
4 (pitch/contact/field/baserun)
Simulation stages
Real MLB ranges
Calibration
Deterministic simulation engine Playwright visual regression
Let's talk
Tell me about your business.
Fifteen minutes, no pitch deck. Tell me what's broken and I'll tell you honestly whether I can fix it.