Writing about how enterprises are rebuilt around intelligence.
I build and write about enterprise AI. Most of what I publish comes out of the work itself. What actually happened when intelligence met a real operation, not what the slide deck promised.
Since January 2024 that has meant agentic AI inside 32 projects across 8 industries. I started working at 7, started building companies at 18, and have gone broke and come back more than once.
Essays on enterprise AI and operational intelligence, written from inside real deployments. What the operation looked like before, what the system actually changed, and what it cost to get there, including the parts that did not work.
Everyone wants agents. Then you tell one to handle a real process and the question appears. How? The answer lives in people, in judgment, in exceptions, and it was never written down anywhere.
You cannot describe a banana to someone who has never eaten one. What the other person understood is the communication, so find out what they have eaten first.
AI's first effect was not on the enterprise. It was on the cost of building software, and then on software's ability to do the work itself.
Gracián wrote in 1647 that the world had peaked and the old rules no longer worked. Every generation since has said it. The question is who crosses.
Everyone wants something, so wanting proves nothing. The only evidence is the price you paid, and whether you agreed to it before you started.
Get the next essay by email.
Roughly one every two weeks. No other email, ever. Unsubscribe in one click.
If something here is useful, wrong, or worth arguing about, tell me.
