Head of GPU & AI Data Center Ops — Remote

Programming.com; United States; Remote; USD 200,000 - 300,000; Full time

Benefits offered by this job

Remote work

Job summary

Programming.com is seeking a Head of Data Center Operations for GPU and AI infrastructure. You will lead end-to-end operations of large-scale GPU clusters, HPC environments, and data center sites, driving reliability, capacity planning, and strategic expansion while working remotely with travel to data centers as needed.

You will manage 24/7 operations teams, collaborate with hardware vendors and engineering, implement automation and monitoring with Slurm, Kubernetes, and Prometheus, and own

Qualifications

• 15+ years in infra engineering or data center ops.
• 7+ years of leadership in production infra.
• Direct experience with NVIDIA GPU clusters is mandatory.

Responsibilities

• Lead end-to-end operations of large-scale GPU clusters and AI infra.
• Manage GPU server deployment, monitoring, maintenance, and lifecycle.
• Ensure high availability and performance for AI training/inference workloads.
• Oversee multi-site data center operations and capacity planning.
• Lead incident management, RCAs, governance, and SLA/SLO performance.
• Build and manage 24/7 ops teams and site leaders.
• Drive automation and fleet monitoring for efficiency and scaling.
• Collaborate with engineering, hardware vendors, and executive leadership.

Skills

Infra leadership

Data center ops

GPU HPC

Leadership

Tools

Slurm

Kubernetes

NVIDIA DCGM

Prometheus

Grafana

InfiniBand

RDMA

Job description

Programming.com is seeking a Head of Data Center Operations for GPU and AI infrastructure. You will lead end-to-end operations of large-scale GPU clusters, HPC environments, and data center sites, driving reliability, capacity planning, and strategic expansion while working remotely with travel to data centers as needed.

You will manage 24/7 operations teams, collaborate with hardware vendors and engineering, implement automation and monitoring with Slurm, Kubernetes, and Prometheus, and own