Writing on AI Governance, Agentic Safety Architecture, Design System Migrations, and Performance Engineering.
Automating a bug pipeline is the easy part. Deciding which tier each bug belongs to is the bottleneck, and it is where the real engineering lives.
A controlled ablation, 42 runs: a behavioral-payload skill added zero marginal defect detection over a no-guidance control. A skill whose payload is documentation inherits documentation's 0%.
Unrestricted shell access handed to an AI agent, then the defense-in-depth architecture that made it survivable: layered hooks, and the self-approval bypass patterns they had to close.
T7,798 generated components in one package. The real cost of the barrel file is not bundle size (it tree-shakes fine), it is build-time parse, plus a dual type-resolution trap that runs but will not type-check.
Two design systems on one page usually means a rewrite. A two-tier CSS architecture, scoped plus utilities, lets them coexist without one.
A migration guide starts rotting the moment it is written. Separate knowledge, procedure, and data, and the guide becomes a product that stays right.
A controlled ablation across nine tasks: a fully-specified CLAUDE.md changed agent behavior 0% of the time, identical to having no rules at all. The data, and why documentation cannot govern behavior.
A 46KB animation library for twelve shapes. The build-vs-buy math that said write the tiny requestAnimationFrame engine instead, and when that call is the right one.
The obvious way to morph an SVG shape tears it apart. Why it breaks, and the least-squares transform fit that makes the animation actually work.
Most migrations are a person reading a guide and hoping. This is the engine that automates it: five-factor confidence scoring, ratcheted state transitions, and gates that never roll backward.
