SIG API Machinery and AI - What Comes Next?

Explore the intersection of Kubernetes API machinery and artificial intelligence in this 30-minute conference talk from KubeCon + CloudNativeCon. Discover how machine learning workloads, from massive training jobs to complex inference pipelines, are pushing Kubernetes APIs to their limits and driving innovation in cloud native computing. Learn about SIG API Machinery's current initiatives and future roadmap as maintainers Joe Betz from Google and David Eads from Red Hat discuss critical AI/ML challenges including scaling API performance for high-throughput scenarios, evolving Custom Resource Definitions (CRDs) to handle complex machine learning workflows and specialized hardware like TPUs and GPUs, and adapting API patterns for efficient processing of large datasets and distributed ML jobs. Gain insights into how to influence design decisions, upcoming Kubernetes Enhancement Proposals (KEPs), and architectural considerations that will shape Kubernetes as a leading platform for AI/ML workloads.