Colophon

I sit at the intersection of product design, systems building, and AI strategy. Over the past few years I've moved from designing complex products and internal tools into defining the operating models, AI-integration strategies, and executive-facing systems behind them.

Today I work embedded with Assurant's executive strategy function — which means I'm often the person who turns an ambiguous leadership need into something real: a working tool, a strategic framework, a patent-worthy program, an executive presentation. Recent shipping has spanned an AI-managed organizational wiki, an AI-planned leadership reporting layer, and a co-invented patent-pending operating model for distributed expertise in regulated environments.

I also write — a monthly internal AI newsletter for busy colleagues, and longer essays on the public side about how applied AI actually lands in regulated industries.

Eight years in the US Army before everything else. Three of those as a Cavalry Scout — observation teams in Iraq and Germany, mounted and dismounted, gathering the information someone else was going to make a decision on. Three more in PSYOP reserves, trained on the psychology of influence and how communication actually moves people. The throughline that survived: how to make decisions before all the information arrives.

Four years as a UX designer in financial services after that — first at American Financial & Automotive Services in Texas, supporting business development tools and the early patterns that became their design system, then into platforms supporting dealer, agent, and operations workflows on multi-million-contract portfolios.

Six years at Assurant. Started as Senior UX Designer on the automotive warranty side, then promoted into Manager, Innovation Research & Design — where scope has progressively expanded from execution into owning AI strategy, early-stage product direction, and operating-model design for the kind of ambiguous, high-stakes initiatives that don't yet have a playbook.

Research first, prototype fast, write the decision down, ship through real users. Translated:

The work starts with understanding the environment well enough to know what context the model actually needs — not what context is easy to gather. Most “AI doesn't work for us” conversations are actually “we never figured out what the problem is” conversations.

From there, prototyping is loud and disposable. I'd rather show a wrong prototype to a real user on Tuesday than ship the right idea three months later. AI-assisted prototyping cut concept-to-prototype time roughly 10× on my team — the value wasn't the speed itself, it was being able to make ten passes where teams used to make one.

Every decision gets written down. Not in a doc no one reads — in the form of a clear hypothesis, a measurable signal, and a date by which we'll know whether it survived. Eight years in the Army taught me that decisions made and forgotten are worse than decisions made and reviewed.

And finally: ship through real users. Internal tools especially. The org that runs the system is the org that gets the feedback loop — anything less is theatre.