Explore the potential of distributed and cloud-native data science workflows using Ray on Kubernetes in this informative conference talk. Discover how Ray, an open-source parallel computing platform, combines the scale-out cluster model of tools like Spark and Flink with a lightweight, scale-to-zero serverless style workflow designed for modern container platforms in the Kubernetes ecosystem. Learn about Ray's comprehensive toolkit supporting various data science and DevOps activities, including ETL, feature extraction, model training, ML pipelines, and serverless inferencing. Gain insights into deploying Ray on Kubernetes and integrating it with JupyterHub to create powerful distributed workflows. Watch a demonstration of Ray in action, showcasing an end-to-end data science project running on Kubernetes. Acquire the knowledge needed to leverage Ray's capabilities for cloud-native data science and enhance your workflow efficiency.