Nine Seconds to Erase the Company
An autonomous coding agent erased a company's entire database in nine seconds. Every customer-accessible backup went with it. The failure was not the model. The failure was the absence of boundaries. This is what 'autonomy' costs when architecture is something a system inherits by omission.
Governed ADRs in the Age of AI Architecture
Architecture Decision Records were a once-a-quarter ritual when humans made architectural decisions slowly. With agents proposing structural choices in every PR, ADRs need to become a daily filter — owned, contested, and signed — or architecture happens by accident.
Intentionality Is Not a Step. It Is a Gate.
Most teams treat intentionality as a checklist item: write the spec, then generate. That framing fails the first time someone is in a hurry. Intentionality only works as a principle when it is wired in as a gate — a generation that cannot proceed without a declared, signed intent.
The Fastest Team Defines Before It Generates
A Claude Code skill collection that interviews you with sixteen to fifty questions before writing a single line of code crossed 20,000 stars on GitHub in weeks. The signal is not novelty — it is the market admitting what fast teams already knew: speed lives in the definition, not in the generation.
August 2026: The Rest of the EU AI Act Goes Live. Is Your Generated Code Ready?
On August 2, 2026, the bulk of the EU AI Act's obligations take effect. For teams that have been generating code with AI for two years without trails of intent, context, and decision, the audit will not ask for the code. It will ask how the code was decided. There are 95 days to be ready.
The Engineer Who Loses Their Own Codebase
In communities where engineers talk to each other about AI coding tools, the same admission keeps surfacing: I am slowly losing grip on my own codebase. The cost of velocity is not burnout. It is the quiet trade of ownership for output. The career that survives the next five years belongs to engineers who refuse the trade.
Signal over noise.