The honest practitioner's take on agentic AI on Kubernetes | BRK222
A practitioner walks through what agentic AI workloads actually require to run on Kubernetes.
“This session is for people already running Kubernetes who are trying to figure out what AI workloads actually need on top of Kubernetes.”
In a Microsoft Build session, Kubernetes veteran Lachlan Evenson gives a practitioner's overview of running agentic AI systems on Kubernetes, covering what works today, what's harder than it looks, and where the ecosystem is heading. The captured content is mostly the talk's intro and framing, with promised demos, customer cases, and repos, so it signals growing interest in Kubernetes as an AI substrate but contains no concrete announcement.