Agentic workloads are turning Kubernetes into an inference and orchestration layer for AI infrastructure.
As AI spending shifts from training to serving models and running agents, the Kubernetes community is building tools to handle huge volumes of inference calls from single prompts.