> _
I'm Bo.
Platform engineer making AI as reliable as infrastructure. I write about agents, systems thinking, and the factory floor where code gets made.
What I Work On
The Problem
Every agent session starts from zero. Your agents forget what worked, repeat what failed, and can't compound knowledge.
The Thesis
The moat isn't the model. It's the bookkeeping — capturing byproducts and feeding them forward so every session starts smarter.
AgentOps
The plugin that does it. Hooks into Claude Code, Codex, and OpenCode to handle the bookkeeping for you. 240+ stars.
Recent Writing
Convergent Architecture: Three Domains, One Principle
→Infrastructure, software architecture, and agent orchestration all arrive at the same answer independently. The intelligence lives in accumulated context, not individual execution.
AgentOps: It's Not Agent Orchestration. It's Context Orchestration.
→Everyone, including me, has been calling it agent orchestration. After months of shipping code with AI agents, I realized we've been naming the wrong thing. This is what context orchestration actually looks like.
The Knowledge Flywheel: How AI Teams Compound Intelligence
→Every AI session generates learnings. Most teams throw them away. The flywheel captures, promotes, and re-injects knowledge so each session starts smarter than the last.
Builds
vibe-check
→A CLI that analyzes git history to give you 5 core metrics—like damage meters but for coding sessions.
Dify Agent Platform
→A multi-tenant AI marketplace running on top secret OpenShift, deployed in a single day.
Open to conversations about AI delivery, platform engineering, and reliability in hard environments.