GPU SRE — Production Linux & HPC Clusters (Night Shift)

hr-pod-hiring-talent-globally; Lahore; On-site; PKR 2,400,000 - 4,200,000; Full time

Job summary

hr-pod-hiring-talent-globally is seeking a senior SRE/engineer to own production Linux GPU clusters, drivers, and high-speed networking. You will troubleshoot complex distributed systems, multi-GPU paths, and NCCL performance, while managing workloads with Kubernetes or Slurm.

You will automate provisioning, image deployment, and remediation using Ansible and IaC, and build observability with Grafana/Prometheus. Collaboration with Platform teams drives reliability and capacity planning.

Qualifications

• 5+ years in systems, infrastructure, or SRE managing production systems.
• Deep Linux troubleshooting with Ubuntu experience.
• Hands-on GPU server operations in production and driver issues.
• Practical network troubleshooting incl. physical-layer faults.
• Programming in Python or a similar language.
• IaC experience with Ansible and Terraform.
• Observability skills using Grafana and Prometheus.
• Bachelor's degree in CS or an equivalent field.
• Experience operating GPU clusters or AI infra in production.
• Production experience with Kubernetes or Slurm; both is a bonus.
• Background in HPC or research computing.
• Familiarity with NVIDIA GPU stack, InfiniBand/RDMA, NCCL.
• Experience with Claude Code or AI coding agents.
• Contributions to open-source projects in cloud-native/HPC/AI infra.

Responsibilities

• Own reliability, availability, and performance of production Linux GPU clusters.
• Lead end-to-end troubleshooting of distributed systems, GPU nodes, networking, and storage.
• Troubleshoot GPU rail and NCCL performance across multi-GPU/multi-node paths.
• Diagnose network faults end to end, including cabling and routing issues.
• Configure and maintain workload managers (Kubernetes/Slurm) and identity/storage services.
• Build automation using Python or Go to reduce toil and improve runbooks.
• Use AI tools like Claude to accelerate automation and incident analysis.
• Automate provisioning, image deployment, and configuration with IaC.
• Design and operate observability with Grafana, Prometheus, Loki; tune alerts.
• Lead on-call, postmortems, and reliability improvements to maintain SLAs.
• Collaborate with Platform and Systems teams on capacity planning and rollouts.

Skills

GPU clusters

Linux administration

Kubernetes

Slurm

NVIDIA NCCL

IaC (Ansible Terraform)

Observability (Grafana Prometheus)

Python/Go

Networking

Education

Bachelor's degree in CS or related field

Tools

Ansible

Terraform

NVIDIA drivers

CUDA

Job description

hr-pod-hiring-talent-globally is seeking a senior SRE/engineer to own production Linux GPU clusters, drivers, and high-speed networking. You will troubleshoot complex distributed systems, multi-GPU paths, and NCCL performance, while managing workloads with Kubernetes or Slurm.

You will automate provisioning, image deployment, and remediation using Ansible and IaC, and build observability with Grafana/Prometheus. Collaboration with Platform teams drives reliability and capacity planning.