Description

You will own the design, validation, and operational maturity of large-scale GPU and AI infrastructure for enterprise partners.

This role is remote.

Responsibilities

• Identify root causes of production issues and prototype solutions under production-like workloads.
• Help partners prepare operational models for new platforms, services, and capacity expansions.
• Analyze operational metrics including incident frequency, recovery time, infrastructure utilization, and cost efficiency.
• Create reusable ecosystem capabilities such as SOPs, reference architectures, assessment frameworks, automation solutions, and agentic workflows.
• Feed recurring pattern insights back to product, engineering, and support teams.

Required Skills

• 8+ years of experience in infrastructure architecture or site reliability engineering.
• Deep expertise in GPU Infrastructure: DCGM, BMC/Redfish, Firmware & Driver Lifecycle Management.
• Networking proficiency with InfiniBand, High-Speed Ethernet, NCCL, and UFM.
• High-Performance Storage knowledge: Lustre, IBM Storage Scale (GPFS), WEKA, or VAST Data.
• Platform engineering with Kubernetes, Slurm, GPU Scheduling, and Multi-Tenancy Architecture.
• Observability tools: Prometheus, Grafana, and OpenTelemetry.
• Automation and IaC: Terraform, Ansible, Argo CD.
• OS and Scripting: Linux Administration, Python, Bash.

Preferred Skills

• Experience supporting large-scale GPU or AI infrastructure at a specialist level.
• Strong root-cause analysis and customer-facing communication skills.

Key Skills

Education

Bachelor's degree