Talent.com
Slurm Administration & Systems Architecture

Slurm Administration & Systems Architecture

MidjourneyHayward, CA, US
job_description.job_card.30_days_ago
serp_jobs.job_preview.job_type
  • serp_jobs.job_card.full_time
job_description.job_card.job_description

Overview

We are seeking a highly skilled HPC / AI / ML Cluster Engineer to support the design, deployment, and ongoing operations of large-scale HPC environments powered by Slurm. This role centers on cluster engineering, administration, and performance optimization, with emphasis on GPU-accelerated computing, advanced networking, and workload scheduling. In this role, you will work closely with our researchers, vendors, and partners to manage Slurm clusters that are used for AI / ML workloads.

Responsibilities

Cluster Engineering & Deployment

  • Participate in the design and bring-up of bare metal HPC / AI / ML environments
  • Architect compute node definitions (NUMA, GRES GPU topologies, CPU pinning) and Slurm partitioning strategies for diverse workloads.
  • Integrate heterogeneous hardware platforms into cohesive scheduling environments.
  • Develop provisioning and imaging workflows (Ansible, MAAS, cloud-init, CI / CD pipelines) for reproducible cluster build-out.
  • Coordinate communications between vendors, researchers, and other partners during cluster bring-up and operation.

Slurm Management

  • Configure and operate the Slurm Workload Manager.
  • Build custom Slurm plugins and scripts (epilog / prolog, pam_slurm_adopt) to extend functionality and integrate with authentication, and monitoring.
  • Manage federated Slurm setups across multi-site or hybrid cloud environments.

    • System Administration & Monitoring

  • Administer Linux HPC environments, including network configuration, storage integration, and kernel tuning for HPC workloads.
  • Deploy and maintain observability stacks for system health, GPU metrics, and job monitoring.
  • Automate failure detection, node health checks, and job cleanup to ensure high uptime and reliability.
  • Manage security and access control (LDAP / SSSD, VPN, PAM, SSH session auditing).

    • User & Stakeholder Support

  • Assist cluster users with developing workflows that make efficient use of compute resources.
  • Containerize HPC applications with Docker / Podman / Enroot-Pyxis and integrate GPU-aware runtimes into Slurm jobs.
  • Automate cost accounting and cluster usage reporting.

    • Qualifications

  • 7+ years experience in HPC cluster administration and engineering, with deep knowledge of Slurm.
  • Familiarity with common AI / ML software package dependencies and workflows
  • Expert in Slurm configuration, partition design, QoS / preemption policies, and GRES GPU scheduling.
  • Strong background in Linux system administration, networking, and performance tuning for HPC environments.
  • Hands-on experience with parallel file system, advanced networking (InfiniBand, RoCE, 100 / 200 GbE), and monitoring stacks.
  • Proficient with automation tools (Ansible, Terraform, CI / CD pipelines) and version control.
  • Demonstrated ability to operate GPU-accelerated clusters at scale.
    • serp_jobs.job_alerts.create_a_job

    Administration • Hayward, CA, US