Back
Mid-senior metro role with popular infrastructure title and broad requirements increases candidate competition.
Strong specialization in GPU, HPC, and Kubernetes observability limits cross-industry transferability.
Explicit 6-10 years plus many mandatory niche technologies increases filter strictness.
Login to See Your Match Score
Create a free account or log in to unlock your CV match score across:
Build and maintain comprehensive GPU infrastructure observability systems including monitoring, alerting, and metrics pipelines for multi-tenant environments.
Develop and operate production-scale LGTM stack (Loki, Grafana, Tempo, Mimir) and related monitoring services with custom exporters, Kubernetes controllers, and systemd services.
Design and implement secure, scalable monitoring architectures integrating multiple orchestration platforms (Kubernetes, SLURM), enabling efficient resource utilization, cost analytics, and incident response.
6-10 years of infrastructure engineering experience, focusing on observability and production service development; exceptions possible for exceptional candidates.
Proven expertise in NVIDIA GPU observability using DCGM, NVML, nvidia-smi, and understanding GPU-specific failure modes.
Proficient in deploying and managing LGTM stack components (Loki, Grafana, Tempo, Mimir), Prometheus, and long-term storage solutions (Thanos, VictoriaMetrics, Mimir).
Strong programming skills in Go or Rust including custom Prometheus exporter development and Kubernetes controller/operator programming.
Experienced in designing multi-tenant, secure monitoring systems with RBAC and query-time filtering for scalable enterprise observability.
Proficient in managing complex infrastructure combining HPC workload managers (SLURM) with Kubernetes, focusing on AI/ML training and inference workload observability.
Capable of owning end-to-end GPU infrastructure monitoring platforms including metrics architecture, alerting frameworks, and cost/utilization analytics, linking technical decisions to business impact.