Explore automated machine learning performance evaluation in this 26-minute conference talk from KubeCon + CloudNativeCon North America 2021. Dive into the intricacies of benchmarking deployed production machine learning models in cloud native infrastructure. Learn about the theory behind ML model benchmarking, including key parameters like latency, throughput, and performance percentiles. Follow a hands-on example using Argo, Kubernetes, and Seldon Core to benchmark a model across multiple parameters for optimal hardware performance. Gain insights into workflow management, reusability, and best practices for evaluating ML models in various deployment scenarios.