How I got here.

I started in enterprise integration in the late nineties — EDI maps, B2B middleware, translators that turned one company’s purchase order into another company’s sales order without anyone having to retype it. Nobody grows up dreaming about this. But if you watch long enough, you start to see that almost every large business, stripped to its frame, is just a set of messages moving between systems that weren’t designed to talk to each other.

I spent thirty-plus years building that plumbing. I got good at it. I also noticed, year after year, how much human effort was still being spent on the pieces the plumbing couldn’t handle — the exceptions, the edge cases, the messages that needed a person to read and decide. That was the wall. It looked permanent.

Then large language models got good. Not toy-good — production-good. The wall moved. The work that used to require a team of humans reading messages and making small decisions could now be done by a system that costs a few cents per decision and runs at 3 a.m. on a Sunday. And the work required to build such a system had collapsed from a team-year into a week, if you knew what you were doing.

So I pivoted. I started Velocity Point as a one-person company with an explicit thesis: build vertical AI products for small businesses, one at a time, and let each one earn its keep. The thesis did not play out. Velocity Point never had paying customers, and I have paused it. The first product, Second Ring, still answers the phone for service businesses that would otherwise miss the call. Revenue Point was the next idea — reactivating dead leads for companies with a database and no one working it — and it did not take on customers either.

The writing fits in here because every book I’ve published is documentation of something I had to learn to keep going. Infinite Leverage is the memoir of the first eighteen months. Ratchet Learning is the playbook that fell out of it. The Bitcoin books are older, but they came from the same instinct — write the thing you wish had existed when you were trying to understand it yourself.

What I’m optimizing for now: leverage, compounding, and building things that outlast the news cycle. I’d rather run three small businesses for ten years than one big one for three. The ratchet only moves one way if you let it.