Nowhere to Put the Canary: Argo Rollouts + vLLM
Running vLLM on a single GPU with room for two pods, and what that does to progressive delivery when there’s no capacity to spare for a canary.
Running vLLM on a single GPU with room for two pods, and what that does to progressive delivery when there’s no capacity to spare for a canary.
Reflections on how I’ve been using AI as a development partner - what I delegate, what I don’t, and what it’s produced.
An essay reflecting on my time using GenAI as a creative partner for a HGSE course that I was a teaching fellow for.