Your question is Secure Kubernetes Cluster for CI Runners. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
Design a secure Kubernetes cluster for public CI runners with clear isolation and cost controls at NVIDIA.
Explain how you would isolate untrusted jobs, manage runner lifecycle, protect NVIDIA infrastructure, and prevent resource abuse. Cover namespace and node boundaries, pod security, network policies, image and credential controls, resource quotas, autoscaling, observability, and failure recovery. Provide Kubernetes manifests or infrastructure-as-code examples and describe how you would validate the design before production.
