/about
// platform engineer · open-source builder · writer
// what I work on
Day job: IC platform/SRE work in defense autonomy. Reliability, release discipline, and operating systems for autonomy infrastructure. Public writing here stays generalized and privacy-safe.
Side lane: open-source tools that make AI agents reviewable, testable, and operable under the same constraints I bring to production infrastructure.
// what I believe
THE 40% RULE
Never exceed 40% of an AI's context window. Below 40%, performance is linear. Above 40%, failures compound exponentially. The most important rule I've learned from turning AI work into reviewable delivery.
THE FLYWHEEL
Every AI session produces byproducts: decisions, patterns, learnings. Most people let these evaporate. I capture and compound them. Citations track what knowledge actually helped. The system gets smarter with every use. The full picture →
THE BROWNIAN RATCHET
Embrace variance (parallel agents). Filter aggressively (validation gates). Lock progress permanently (merge to main). Chaos in, quality out. The ratchet only turns forward.
The agent is throwaway. The accumulated context is the intelligence.
You can't build until you've cleared what doesn't belong.
Value nobody can see is indistinguishable from value that doesn't exist.
Many short sessions = still deciding. Few long sessions = building.
// what I'm building
Open-source tools for reliable AI delivery. The theme stays the same: make fast AI work reviewable, testable, and operable under real constraints.
AgentOps
→Coding-agent plugin for Claude, GPT, and other models. Skills run research, planning, validation, implementation, and learning extraction. Knowledge compounds across sessions.
Gas Town + gastown-operator
→Contributing to Steve Yegge's context-orchestration ecosystem. How agent workflows become durable operating systems.
Beads
→Git-backed issue tracking that survives agent sessions. Durable workflows that persist across runs and keep coordination explicit.
Case studies and the operating loop are at /builds and /workflow. The methodology has its own home at 12factoragentops.com.
// how to reach me
Building in reliable AI delivery, platform engineering, or agent-driven workflows? I want to hear about it.