PROJECTS
Writing about methodology without building anything is just theory. These are the projects where the ideas get tested — and sometimes proven wrong.
A co-op adventure RPG built with my daughter. 13+ sprints using spec-first methodology — every mechanic started as a written intent document before any code was written.
Under active development — not yet open to the public.
A bilingual (English/Hebrew) family allowance ledger I'm building for my daughter — to solve a real problem in our home, and to teach her about saving, goals, and how interest works. Built spec-first, like everything else.
In active development toward V1.0. Built bilingual with full Hebrew RTL from day one.
A software implementation of original academic research — a weighted model for measuring what actually separates high-performing teams from average ones. Draws on Edmondson's psychological safety research and Google's Project Aristotle.
Completed for B.Sc. degree submission. The research behind it shapes everything I write about teams.
An algorithmic trading bot. 261 S&P 500 stocks, ML ensemble, 93 features. Built without a proper methodology. The cautionary tale — where the absence of a spec showed its full cost when the system drifted silently from original intent.
Paused. The lesson here became the foundation for the spec-first methodology.
More projects in progress. This page grows with the work.
Short AI-powered browser games — no install, no account. The full collection is over there.
Go to Games →