Operations
Measure the thing, not the config
Changing a setting and observing that the setting changed is not a measurement. A short field note on optimisations that did nothing.
ReadBlog
Field notes on agent-built software.
What we have actually measured, including the times the measurement contradicted what we expected. Mostly about evidence, review, and the gap between a passing check and a working system.
Practice
The deploy gate is the product
Every meaningful decision about autonomous software development eventually reduces to one question: who is accountable when it ships?
ReadReview
Who reviews the reviewer?
Independent review is only meaningful if the reviewer can actually disagree. Here is what independence has to mean structurally, not aspirationally.
ReadEvidence
A green build is not evidence
A passing test suite tells you that the checks you wrote agreed with the code you wrote. That is a weaker statement than most teams treat it as.
ReadPut an intent in and see what comes back.
Every run produces a plan, a build, an independent review, and a gate you control.