On-Demand Ray Clusters in ML Workflows via KubeRay and Sematic

Explore the benefits of leveraging short-lived Ray clusters in broader ML workflows through this 13-minute video presentation. Learn how to utilize KubeRay for managing ephemeral Ray clusters on Kubernetes, enhancing reproducibility, efficiency, and observability in distributed training and multi-step data modeling pipelines. Discover insights from Sematic's experience in implementing KubeRay and Ray for single-purpose cluster management. Gain valuable knowledge about improving ML workflows using on-demand Ray clusters, with a focus on practical applications in distributed systems and AI workloads.