For decades I ran water infrastructure from the operator's chair, not the analyst's. I ran a capital program approaching a billion dollars a year at one of the largest water utilities in the country, so I know from the inside why infrastructure money does not move. I inherited and ran a multi-billion-dollar, legislatively mandated program, which means I have lived the politics and the consent decrees, not just read about them. I launched a major metering and modernization initiative. I worked with a global resilience program and an international engineering firm on a city water resilience framework, and I took part in a congressionally directed national pipeline study that drew on field data from hundreds of utilities.
Across all of it, the wall was always the same. The problem was never just old pipes. It was fragmentation, all the way down to the org chart. A utility knows things in a thousand places and connects almost none of them, so the knowledge that could prevent the 2am failure sits one department away from the person who needs it. Then the people who held that knowledge in their heads started to retire, and it walked out the door with them.
So I did the unglamorous thing. I went and learned the AI and the building, hands on. Not to chase the hype, because pouring AI on fragmented data just gives you a faster mess. I learned it to fix the data first, and to connect the scattered pieces into something a utility can actually trust. Adapt, do not pivot. The thirty-year operator still matters. He just has new tools and a shorter clock.
Think of a utility the way you think of a body. It already has a nervous system in its sensors and SCADA, firing signals all day. What it has never had is a brain that connects those signals, remembers what it learned, and tells you what to do before the fever becomes an emergency. A float switch in a wet well senses a rising level and turns a pump on. That is the simplest version of the idea. The work of my second life is that same loop, scaled to a whole system, with the reasoning attached so you can trust the answer.
Today I run a small, bootstrapped AI team that ships like a company many times its size. This site is where the rest of the work lives: the essays in The Systems Lens, the OneWater Lexicon, and the OneWater Minute. Every claim here is sourced, and I write the way I talk. I am usually only a couple of steps ahead, and I would rather show you the path than lecture you from the end of it.