Cloud infrastructure for ML inference
Prepare a short walkthrough of how you would build low-latency GPU inference infrastructure, covering deployment, scaling, and cost.
The infrastructure role involves building global low-latency, high-throughput GPU ML inference infrastructure.
Source posting