Search
Features
Long reads from work that shipped: trust layers, adversarial review, eval integrity, and production agent systems. Slower than the daily firehose. Denser per page.
The daily blog at /posts/ is a firehose: one entry a day from whatever actually shipped. Features is the other end. Hand-picked long reads that earn the time it takes to finish them.
Featured pieces rotate as the work moves. Older multi-chapter series stay in the library so nothing vanishes from the archive. Written in the same operator voice as the rest of the site: first person, specific, no hype.
Let the Model Judge. Make the Code Decide.
The model owns judgment. Deterministic code owns the gates, the commit, and the audit trail. How I rebuilt a daily writing pipeline so a hung session can no longer brick tomorrow.
Adversarial Review: The Six Lenses That Halted a Rollout
Six lenses, live state, eighteen risks. I thought the safety work was shipped. The review proved three assumptions were theater. The durable rule: the model proposes; the system owns identity, trust, and state.
The Moat Is the Trust Layer
A local-RAG demo became Intent NEXUS. The moat is not the chatbot. It is egress control, code-enforced citations, a ledger you can check, injection defense, and evals that cannot lie quietly to you.
Noise-Robust LLM-Judge Evals: Don't Sign a Coin Flip
An un-seeded LLM judge is a coin flip even at temperature 0. I stopped signing single-call noise into a transparency log. Fold N recorded votes, gate on agreement, label replay fidelity honestly.
Rent the Agent, Own the Proof
Claude Tag is the best agentic Slack teammate you can rent. The question it does not answer: who owns the memory, and who can verify the audit log? What customer-owned, verifiable agent governance looks like in code.
in-progressAgent-Native Mobile Testing
A three-chapter series on plugin authoring, agent reliability on real-device clouds, and AI triage for agent-orchestrated mobile cloud testing: worked through kobiton/automate as the running case study.
in-progressBuilding MCP Servers in Production
A three-chapter case study on shipping production-grade Model Context Protocol servers: synthesized from the daily case-study posts that documented the work.