Projects

What each AI coding model actually costs per passing test.

27k lines of Rust with 532 tests, released through crates.io, a Homebrew tap and a Scoop bucket from one CI pipeline, across sixteen releases.

Six unlike usage logs, a SQLite database, tailed JSONL sessions and a telemetry file, normalised into one priced model where an unknown rate stays unknown instead of becoming $0.00.

Measures cost per passing test per model, so routing and escalation decisions rest on delivered results rather than token counts.

A terminal markdown reader with a contents pane that knows where you are.

A 27k-line Rust TUI with 897 tests and sixteen releases: a markdown renderer written against a cell grid, packaged for crates.io, Homebrew and Scoop, with a nixpkgs submission in review.

Turns "something looks off about this run" into an evidenced answer.

33k lines of TypeScript and React 19 with 2,682 test assertions, Playwright end-to-end runs that include axe accessibility checks, a Web Worker analysis path and a build-time Content Security Policy.

Diffs two pipeline exports record by record, separates systemic field loss from scattered drift, and recovers only what a per-source policy allows, with provenance on every value.

Sixty tech sources, scraped concurrently, deduplicated three ways, served from SQLite.

An async Python 3.12 service on FastAPI and httpx with 108 tests, a Dockerfile, a health check, and a Fly.io deployment that refreshes itself every twenty minutes.

Sixty sources fetched concurrently behind a semaphore with backoff, three-pass dedupe (canonical URL, title fingerprint, Jaccard above 0.85), idempotent WAL upserts, FTS5 search and keyset pagination.

Near-duplicate detection by token-set Jaccard similarity inside a two-day window, and weighted keyword scoring across thirteen categories, with the thresholds written down and tested.

A terminal anime client, forked with its licence honoured.

430 commits on a GPL fork that keeps its attribution straight, with a Windows installer and AniList sync.

A job-postings pipeline with a data-QA core: every rejected row says why.

Adapters split fetch from pure parse so 44 tests run offline on saved responses; the UI starts runs as subprocesses, so a failing adapter cannot take the dashboard down.

Nine auditable checks reject a listing with the check name and reason attached; stable ids per source and first-seen / last-seen timestamps make dedup across runs a fact, not a guess.

A math skill tree that tunes its own difficulty with Elo.

A TypeScript game with an adaptive-difficulty loop, 113 commits in.

Per-skill Elo ratings and a target success rate choose the next problem; spaced repetition schedules the review.

Breakout in C++20 and raylib, with a boss fight.

C++20 and raylib: a procedurally generated campaign, a boss fight, doctest tests, and no runtime to hide behind.

A roguelike deckbuilder where solving algebra is the combat.

Lua and LÖVE: a rules engine that checks algebraic equivalence on every card play, under a scoring system built to reward the longer solve.