Blog
Notes on governed autonomy.
Writing on agent governance, receipts, bring-your-own-engine architecture, and what it takes to trust AI with real work.
Four silent failures in our own fleet, and what a receipt would have caught
We pointed our agent fleet at our own business portfolio and found four production paths that had been reporting success while producing nothing, one of them for eighteen days.
Twenty-four watch passes on agentic commerce: what a research loop found
A scheduled research agent read one market four times a day for a month. The consumer-facing agentic checkout story is retrenching while the payment rails underneath it consolidate.
What are receipts for AI agents, and why audits beat vibes
A receipt asserts the outcome an AI agent produced, not just that a tool returned 200. Here is the shape, the contrast with transcripts, and one worked example.
Bring your own engine: why your agent runtime should be a commodity
The governance around your agents is the hard part. The runtime is interchangeable. Two engines already prove it, and MCP slots in any other.