Dispatches from O'Reilly: The right amount of spec for agentic development
When code gets cheap, the hard part is deciding what “correct” means and building a reliable way to check it.

Articles in the “Dispatches from O’Reilly” series.
When code gets cheap, the hard part is deciding what “correct” means and building a reliable way to check it.

Your semantic layer is a risk mitigation strategy. Not risk in the abstract, compliance-framework sense, but the practical, operational risk that quietly drains organizations every day.

Experiments in agentic engineering and AI-driven development.

Selective control in autonomous AI systems: Why governing every decision breaks autonomy—and how runtime control actually works at scale.
