Every generation is handed a force it does not yet know how to hold. My grandparents were given electricity — a thing that could light a city or stop a heart — and they learned to put it behind fuses and breakers until lightning became the quiet hum behind a switch. My parents grew up as the engine remade the world, and we answered that power with brakes and roads and rules. I was raised at the dawn of the computer — an Atari, an Apple II, the small miracle of a paddle chasing a square of light across a screen, and FORTRAN as a required course for every engineering student at Berkeley. And now a new fire is here. It is beginning to think back.
The question everyone is asking is the oldest one we have: can we harness it? My whole life keeps pointing at the same answer — carefully, with a hand on the wheel.
Numbers, then buildings
My career started in numbers. At eighteen I went to work in the accounting department at Turner Construction in downtown Los Angeles, walking checks to the bank for some of the biggest towers of the era. Numbers led to buildings: a civil-engineering degree from Berkeley, years as an outside construction sales rep calling on homebuilder design centers across Arizona and Northern Los Angeles, and on architects when I carried a commercial flooring line into the specification world. I co-founded Strata Development and took Santa Fe Palms from a rezoning all the way to entitled, shovel-ready drawings. I spent years in design centers — flooring, cabinets, countertops — on projects with real names. And I added the finance lens for good with the Wharton Online real-estate certificate: financial modeling, the capital stack, and the institutional-grade memos that turn a hunch into a defensible case. The thread was always the same — take something complicated, organize it, and make an honest decision.
The turn
For a long time “AI,” to me and to almost everyone, meant a chatbot — a clever question-answerer that waited for us to speak. Then, in late 2025, the ground moved. The machines learned to do more than answer: to take a small, careful step, check the work, and take another — with a person nearby, steering. The “chat-bot” era became the “agent-first,” agentic era, and the first time I watched an agent carry real, multi-step work end to end, I was all in. Emphatically.
Enthusiasm wasn’t enough, though, so I found a method: the A.G.E.N.T. framework from Dr. Ulla Kruhse-Lehtonen and Dirk Hofmann of DAIN Studios. It gave me a disciplined way to ask the right question — not “where can we bolt on AI?” but “where is this organization actually ready for agents, and how do we redesign the work around the people, the goals, and real efficiency?” That is the work I do now. And I build — agentic construction cost models are one early example of putting the method to work on a problem I know cold (estimates are approximate and not a guarantee).
How I want to build
I believe we can harness this fire the same way we harnessed every one before it: with humans in the loop, guardrails built on purpose, and a knowledge base you can trust underneath. Powerful tools stay kind and honest when compliance is built into the foundation — when the rules of the world are watched, the work is checked, and a person keeps a hand on the wheel. That is the future I’m interested in helping make: organized, careful, and generous. There’s an old word for a tool that becomes part of how you think — mindware — and that, more than anything, is what this has become for me: not a replacement for thinking, but a companion to it.
I’d rather earn your trust than sell you something. The tools finally caught up to the ambition. Honestly — I’m just getting started.
More from my agentic journey
A few companion pieces, written plainly and offered to be useful:
• What About AI? — the two-week deadline and the question that started it all.
• A Confused Mind Says No — why most resistance to AI is confusion, not fear, and how we get past it.