product-minded engineer building ai products β agent harnesses, computer use, and evals
- agent harnesses & dev tools β building the run loops, tooling, and interfaces that make ai agents actually useful
- computer use & evals β giving agents hands to operate real apps, and rigorous ways to measure what works
- product sense, 0β1 β end-to-end ownership from concept and design through architecture to polished ux
| project | description |
|---|---|
| pi-gui | desktop harness for the pi coding agent β electron + react, real-time streaming timeline with inline diffs, worktrees, and published npm extensions (goal mode + subagents) |
| computer-use-mcp | agent-agnostic computer use as an mcp server β a signed swift binary that lets any agent see and operate mac apps in the background, without stealing cursor or focus |
| openbench | eval harness comparing coding-agent harnesses β same model, harness varies: how much does the scaffolding around a model actually matter? checker-graded, reproducible |
| openci | package manager and discovery platform for ai-agent ci workflows β find, install, and maintain github actions from any repo |
| release bots Β· @codexreleases Β· @pichangelog | ai changelog bots watching github releases and auto-drafting threads with the claude api β ~20k followers combined |
| pluto | 0β1 trading platform on hyperliquid β perps, spot, cross-chain deposits |
typescript python go react swift
π Β· @codexreleases Β· @pichangelog Β· linkedin Β· github