- 14 Projects
- 15 Posts
- 9 Live demos
- ~20 Agent workers
The Notebook
Recent writing
-
Trust is a property of the system, not the agent
agents
A near-miss with a git flag once put three volumes of real data one sync cycle from deletion — caught by luck, not by design. The right fix wasn't more vigilance. It was building the trust into the system itself: an agent doesn't need to be trusted, an action needs to be cheap to undo.
-
What breaks when you run twenty agents at once
agents
A failure taxonomy from a year of running coding-agent fleets unattended. Eight ways a fleet dies, none of which involve the model — and the operational discipline they add up to.
-
From idea to agent fleet: how I build applications with LLM coding agents
llm-agents Last verified ·
The end-to-end workflow I use to turn an idea into working software with autonomous LLM agents: a docs-first repo scaffold, one reviewed plan, a dependency-aware work queue, and a fleet of workers that process it — plus the failure modes I hit for real.
The Constellation
The Arcade
Live demos
- Adaptive Auto-Battler — PvE auto-battler with a self-learning enemy AI — runs entirely in the browser.
- Face Detection — Real-time face detection in the browser via webcam.
- Gait Analysis — Real-time human pose and motion tracking from a webcam feed.