Thinking in public
I write about AI architecture through the systems I build and the frameworks I read. The focus is practical: agent permissions, design choices and the evidence a decision needs.
These are the full pieces. Excerpts published elsewhere link back here, so the version on this domain stays the one of record.
Two proofs, or it does not exist
Most teams can watch their agents run. Far fewer can say whether they work. This is the ledger I built so that no mechanism in a live agentic system is allowed to claim it works without evidence that it fired — including the day the gate caught itself.
You cannot motivate an agent
Managing agents and managing people are different jobs with opposite failure modes. The leadership layer over agentic work is thinner than anyone admits — and thin is not the same as absent. What survives the thinning is the part that was always hardest.
What running engineering with agents taught me about running it with people
An agent that produces work you have to check is a junior engineer, structurally. What running an engineering function that way taught me about specification, delegated authority, and where the analogy stops.
Memory that agents can read but not quietly rewrite
Give an agent memory and you give it two things: recall, which you wanted, and a channel through which its own past output returns as input wearing the same clothes as everything else.
Reading Singapore's agentic AI framework as an architect
Most people read a governance framework to check whether they comply. I read Singapore's agentic AI framework for the sentences that would change what I build. There are four, and only one is about ethics.
Someone always had to sign
Working on Singapore's interbank settlement system taught me that nobody ever asked what the system could do. They asked who approved it, on what evidence, and whether we could show them two years later.