Engineering Principles

Observability for AI-Assisted Development

Developers using AI estimate they’re 24% faster. A randomised controlled trial measured them at 19% slower.That’s from METR’s 2025 study. These were experienced open-source developers working on their own codebases with tools they chose. Their self-assessment was off by over 40 percentage points.If your perception of AI’s impact is that unreliable, what are you actually measuring?If you didn’t measure before AI, measuring with AI won’t work.You can’t attribute improvements to AI if you don’t know what “before” looked like. Cycle...

Read

The Judgement Pyramid: Reasoning vs Measurement

A team I was talking to had built a code-review skill in Claude Code. It looked good: a careful prompt, a set of rules, and examples. Then it shipped a PR with a nested recursive loop inside another recursive loop. Cyclomatic complexity any static analyser (PMD, radon, ESLint’s complexity rule) would have flagged in milliseconds. The skill missed it.The team’s response was to iterate the skill. More examples. More rules. A few more lines of prompt. The context window kept...

Read

The Comfort Plateau AI Built For You

I’ve spent the last year watching engineers and writers adopt these tools, and the same pattern keeps showing up. The first three months look transformative. The next nine look almost identical to the first three.The output keeps coming. The confidence keeps building. The skill stops moving.This isn’t a story about AI making people dumber. It’s a story about a comfort plateau that didn’t exist before, and what it costs to stay on it.Draw a spider chart. Hundreds of axes: SQL,...

Read