// Insights

Writing, builds, and experiments

Essays drawn from leading AI work, and the personal projects I build on my own time. Personal projects are labeled as such; several include interactive illustrations.

Writing

August 23, 20269 min read
From a Team of One to a Defense AI Program: Building ORBIS's AI Department
I founded ORBIS's AI department as its only engineer. Five years later it was a core team of ten inside programs of 30+ people spanning company departments, contract experts, the U.S. Navy, shipyards, a university, and industry partners. The hurdles, the results, and the servant-leadership model that held it together.
August 1, 20267 min read
What Shipping AI Into Submarine Construction Taught Me About Production ML
Most ML projects fail in the distance between a validated notebook and a system running inside someone else's workflow. What years of building AI for Department of Defense inspection pipelines taught me about that distance.
May 20, 20266 min read
Running LLMs Where the Data Can't Leave the Building
For defense, healthcare, legal, and finance workloads, sending data to a commercial API is often prohibited outright. What actually works when the models have to run inside the boundary.

Builds & experiments

Personal projects, built at home to understand how things work: sensors, radios, robots, and local AI. None of it is employer work, and several are still in progress.

Also available by RSS.

Writing, builds, and experiments

Writing

From a Team of One to a Defense AI Program: Building ORBIS's AI Department

What Shipping AI Into Submarine Construction Taught Me About Production ML

Running LLMs Where the Data Can't Leave the Building

Builds & experiments