AI Agents

This is part 6 of the Agent production system series (boundaries, not “series over” — later posts dig into GitHub tokens and the landing floor). Previous: Claim safety. Constellation: Human approval, Eval gates, Orchestrator. A series about agents should spend real time on boundaries. Not because automation is bad — because production systems are defined by what they refuse. I will automate aggressively inside the funnel. I will not pretend every human act is a waste waiting for a model. The goal was never maximum automation. The goal is a system that moves work fast and still has a place for a human to say no. ...

This is part 5 of the Agent production system series. Previous: From Xbox SLAs to agent fleets. Constellation nodes: Eval gates, Search, Human approval. Agents are fluent. Fluency is not the same as true. Claim safety is the discipline of keeping numbers, titles, and authorship tied to evidence — especially when a model would rather sound complete than sound correct. It shows up in eval gates, in human approval, and in public artifacts like this site and Resume Builder. ...

This is part 4 of the Agent production system series. Previous: Docker homelab ops. Constellation: the whole agent production map — especially Orchestrator, Eval gates, and Human approval. I am a former Senior Technical Program Manager at Xbox/Microsoft. That sentence is past-tense on purpose — and it still explains how I design agent systems. People hear “TPM” and picture a calendar full of status meetings. At its best, platform TPM work is the opposite of that caricature: critical path, risk, multi-team alignment, and SLA performance under real blast radius. Someone has to know what is actually on the path to “done,” what can kill the ship window, which teams are blocked on which contract, and when a green dashboard is lying. That is the job. ...

This is part 3 of the Agent production system series. Previous: Human approval. Constellation nodes: Docker host, Azure / ADO, Dashboards. Agents need somewhere to live. Chat UIs hide that fact. Production does not. I operate a Dockerized homelab with 20+ production containers split across two physical hosts — a Mac Mini M4 running the compute-heavy services under OrbStack, and a Synology NAS handling bulk storage, VPN-gated download automation, and the public-facing reverse proxy. That number is on the About page for a reason: it is a real ops footprint, not a laptop demo. ...

This is part 2 of the Agent production system series. Previous: Eval gates are not optional theater. Constellation node: Human approval. People hear “agent fleet” and picture a fully lights-out factory. That is not what I run — and not what I want. I run a system where agents move the bulk of the work and a human still owns irreversible decisions. The merge button (and its cousins: force-push, secrets, prod infra, public claims) is not a relic. It is a designed control. ...

This is part 1 of the Agent production system series. Start with the system map if you haven’t read it yet. On the interactive diagram: Eval gates. Demo agents look smart until they touch a real repo. In a demo, the agent writes the code, the code runs once, everyone claps. Nobody checks what happens the second time, or the tenth, or the time the agent decides the fastest way past a failing test is to delete the test. Then you learn the hard lesson: intelligence without a gate is just a faster way to ship a bad change. ...

Most “AI agent” demos stop at a chat box. What I care about is the opposite: a production system with routing, tools, gates, hosts, and a human who still owns irreversible decisions. This post is the written version of the interactive map on my About page — how the fleet is wired, what each layer is for, and what I refuse to automate. The mental model Think factory floor, not chatbot: ...