About

Like many a nerd before me, I got hooked on coding through a TI-83 graphing calculator.

In middle school, those calculators were a convenient loophole for gaming in class. At some point, games like Block Dude stopped cutting it, so I took a creative leap of faith and wrote my own. I picked up TI-BASIC by reading other games' source code and threw together a simple text RPG. My friends loved it and pushed me to write more, and before long I was programming everything. Every new math or physics formula became a program: unit conversions were trivial, and the harder formulas were genuinely fun. By high school I'd written a program that drew a z-axis on the graphing screen and plotted three-dimensional objects. The much more expensive TI-89 (which I begged my parents for at Christmas) did that out of the box, but it still felt like quite the accomplishment. When a programming class showed up in the curriculum my sophomore year, I jumped on it.

Programming, history, and wrestling were my passions through high school. At Santa Clara University I majored in history, because it was what I loved most, and minored in computer engineering to keep my skills sharp and employable. The low-level side was challenging and rewarding: binary arithmetic in embedded systems, and building a 32-bit processor in Verilog for advanced computer architecture. That's where I got a real look at how computers actually work.

After school, I started out in the literal field. After a short stint programming an irrigation pump on a friend's farm in eastern Washington, I interned with a California water management company, programming and installing Campbell Scientific data loggers at freshwater spring sites. The money, it seemed, was in software, so I switched back to Java and JavaScript and spent the next four years in business intelligence, a couple of them in Vienna, Austria. When I came back to the States, I landed a job at Nike, and I've been there ever since.

At Nike I found my niche in cloud performance engineering, and spent almost ten of my eleven years there going deep on AWS and on front-end and back-end performance. Then LLMs arrived, Nike's digital landscape shifted dramatically in a single year, and I moved into SRE automation. After a decade of writing code by hand, the transition to AI has been a welcome one. The technology genuinely excites me, and these days I spend most of my time shepherding my genius robot toddlers, at work and at home.

At work, that's CCE SRE Mission Control, the agent-ready knowledge platform I architected for my Cloud Core Engineering (CCE) site reliability team, and CCE-SRE-Expert, the production Slack agent built on top of it. I also help my teammates work alongside agents day to day, from running parallel agent sessions to keeping a person in the loop for anything that matters.

At home I run Trickle.news, a public digest that pulls from primary sources, like medical and scientific research and the federal and state registers, scores each item for impact and interest, and counts how much news coverage it got. The goal is to surface high-impact stories that get little or no mainstream coverage. I'm also building Ascendant's Archipelago, an original waterpunk world that started as a lore vault and is now an Unreal Engine game built by coding agents, with me as creative director. If you'd like to play a demo, email me and I'll send one over!

All these agent projects have shaped how I build with AI. I let measurement, not trends, set the direction: evals before architecture debates, cost models before adding more agents, and a regression test with every behavior fix.