The Zombie PR Loop: Why My Agents Kept Working After the Job Was Done
Five AI dev workflows kept pushing fixes to PRs that had already merged. The bug wasn't the poll — it was where I checked whether to stop.
● AI-agent engineer · healthcare operator
A real and growing share of my work — and my company's operations — runs on a fleet of AI agents I built. This is the honest log of how it actually works, what it still can't do, and what broke along the way.
Five AI dev workflows kept pushing fixes to PRs that had already merged. The bug wasn't the poll — it was where I checked whether to stop.
When you build an agent or skill to automate a form-heavy workflow, keep the *procedure* and the *sensitive values* in strictly separate places. The reusable…
Filing a ticket creates the appearance of progress while nothing is actually owned, scheduled, or moving.
An LLM's attention is cheap and a human's is scarce. Spend the cheap one to protect the scarce one.
A single web platform that runs a clinic's whole day — check-in, a live provider queue, lightweight charting, telehealth, staff messaging, and billing — instead of a dozen disconnected legacy tools.
The orchestration layer that launches agent sessions, watches them for trouble — rate limits, context exhaustion, dead sessions — and recovers them automatically, using deterministic scanning instead of fragile notifications.
A kiosk and mobile intake flow where patients check themselves in — confirm demographics and insurance, sign policies, answer pre-visit questions — and drop straight into the live clinic queue.
Building agent systems inside a real business — or running a clinic and wondering what this could look like for you? That's exactly the conversation I want to have. No form, no funnel.