Scaling ML Pipelines with KALE - The Kubeflow Automated Pipeline Engine

Explore KALE (Kubeflow Automated Pipeline Engine) in this 25-minute conference talk from KubeCon + CloudNativeCon North America 2021. Learn how to overcome common hurdles in developing AI and deep learning models by designing scalable, real-time data pipelines. Discover how KALE bridges the gap between data scientists' work in Jupyter Notebooks and production-grade pipelines, enabling efficient model training at scale and real-time serving. Gain insights into the best infrastructure setups for experiments and learn how to streamline the process of making models production-ready without requiring extensive infrastructure knowledge from data scientists and engineers.