Start with what happened
Every essay begins with something tangible: a customer conversation, a factory floor, a classroom, a product decision, a city, or a family story. Building and scaling are sources of evidence. I am less interested in polishing the past into a neat lesson than in asking what the work changed—in the product, in the people using it, and in the person doing it.
Technology is part of the subject because it changes what people can do. It is never the whole subject. Customers, buyers, partners, operators, and communities reveal where an idea holds, where it fails, and whether it deserves to last.
Follow the consequences
Across the archive, the recurring questions are practical: What is work for? How is judgment learned? Which institutions endure because they adapt, and which principles should not move? What becomes possible when technology extends human capability without deciding what people ought to value?
My writing stays close to operating details because abstractions don't hold up in the real world. The real world is full of edge cases! Lastingness is less about permanence than care: knowing what to change, what to preserve, and who will live with the result.
Public archive
A Future for Everyone - trust, accountability, and the operating details required to make an AI future credible.
On Intuition - how practice, pattern recognition, and care become judgment.
Steady in the Heartland - what enduring institutions teach about adoption, change, and what stays constant.
Work as Dignity - why technology should extend human capability without erasing the meaning people find in work.
Of Math and Men - using mathematical reasoning when a model can clarify the tradeoff but cannot decide what matters.
Made in 9 Years and 1 Day - the long apprenticeship behind work that looks quick once it is finished.
What If Things Go Right? - optimism as a practical discipline for designing futures people can believe in.
Learning from Machine Learning - what Monte Carlo Tree Search, UCB1, and optimism under uncertainty can teach beyond the model.