Ownership: You are comfortable being the person accountable when the cluster is unhealthy or capacity is tight.
Systems Intuition: You understand how schedulers, containers, networking, storage, and hardware interact. You can reason about failure modes and design systems that degrade predictably.
Operational Rigor: You value observability, reproducibility, and clear operational boundaries. You leave systems in a state that other engineers can understand, operate, and debug without you.
Pragmatism: You can support experimental research workloads without forcing everything into a rigid "production" mold. You know when to stabilize and when to allow controlled chaos to speed up discovery.