Originally published at llmkube.com/blog/operator-fixed-its-own-bug-on-amd. Cross-posted here for the dev.to audience.
(LLMKube is an open-source, Apache-2.0 Kubernetes operator for self-hosted LLM inference across NVIDIA, Apple Silicon, and AMD. Foreman is its agentic harness.)
Last time Foreman built a feature for itself. This is the sequel, and it is a better story: this time it fixed a bug I had just shipped, the model doing the fixing ran on a consumer AMD box on my desk, and the most useful moment in the whole run is where the model got it wrong.
TL;DR
Wiring up Claude Code against one of my own local models surfaced a real bug in the operator: a hardcoded 60-second timeout that silently capped every request regardless of what you configured.
