استخدام Sr. Platform Engineer
شرح موقعیت شغلی
1) K8s cluster-of-clusters, API server, custom CRDs (GPU topology-aware from v0 — vendor, architecture, NVLink gen, NVSwitch presence, AMD MI350X fields), reconciliation loop, upgrade strategy.
2) Terraform/Pulumi modules for AWS (MVP), then GCP/Azure/OCI/neoclouds. Full lifecycle: provision, scale, teardown. gRPC API surface per provider adapter.
3) SLURM operator + K8s-native for MVP. Ph 2 adds Volcano/Kueue with topology-aware placement (NVLink/NVSwitch group affinity for NVIDIA; Infinity Fabric affinity for AMD MI350X).
4) OpenTelemetry instrumentation across all platform components (control plane, schedulers, GPU device plugins, cloud adapters). Prometheus + Grafana backends. NVLink saturation, ECC, MIG utilisation, AMD ROCm SMI counters from MVP
5) Automated GPU/CPU health checks on node provision. Extensible test catalog: link stability, ECC, PCIe errors, thermal. Results stored in Ground Truth Library. GPU Perf Eng consulted on benchmark thresholds
مهارتهای مورد نیاز
- MVP
- platform engineering
- GPU
حداقل سابقه کار
- مهم نیست
جنسیت
- مهم نیست
وضعیت نظام وظیفه
- مهم نیست