Short essays on production AI systems, delivery governance, and the operational layer that makes both work.
Most writing about AI systems is either research-flavored or marketing-flavored. The operational middle is underwritten: evaluation, rollback, audit evidence, ownership, and the delivery controls that keep systems from drifting into liability.
- 001 — What production-grade AI actually means
- 002 — Judgment-as-code: governance you can deploy
- 003 — Bounded autonomy: designing agents that fail well
AI systems do not fail only because a model is weak. They fail because nobody designed the operating envelope around the model.
These notes focus on that envelope.