PMLE › Serving and Scaling Models
This domain covers deploying, optimizing, and operating models on Vertex AI: endpoints, custom containers, batch vs online prediction, autoscaling, GPUs/TPUs, and Vector Search indexes. Questions are scenario-based, asking you to choose the configuration, index type, or rollout strategy that meets latency, throughput, cost, and freshness constraints.
PMLE Serving and Scaling Models — All 168 Questions
Every question in this domain with answers and detailed explanations.