- Build, validate, and benchmark new customer clusters, then hand them over with documentation.
- Debug multinode and NCCL performance issues, GPU and hardware faults, driver and fabric issues, and container and scheduler problems.
- Coordinate provider escalations involving maintenance windows, RMAs, hung nodes, and disputed fault attribution.
- Own named customer accounts, learn their workloads, and advise them on infrastructure changes.
- Own monitoring and alerting for customer clusters using Grafana, Prometheus, or similar tools.
- Automate repeat work and turn escalations into runbooks for the L1 tier.
KubernetesGrafanaPrometheus+2 more