Senior HPC Engineer - Hybrid, GPU & Slurm Expert

LA International; Stevenage; Hybrid; GBP 92,000 - 138,000; Full time

Job summary

LA International seeks an experienced Senior HPC Engineer to support and evolve enterprise HPC platforms used for research, engineering, and computational workloads on a long-term program. The role requires hands-on Linux (RHEL), Slurm, and Nvidia GPU-based environments, including DGX/HGX systems.

The successful candidate will administer, optimize, and troubleshoot production HPC clusters, collaborating with researchers and engineers to deliver infrastructure improvements.

Qualifications

• 5+ years of experience supporting HPC environments.
• Strong admin experience with Red Hat Enterprise Linux (RHEL 7/8/9).
• Hands-on Slurm in production.
• Experience managing HPC clusters and scheduling infrastructure.
• Experience Nvidia GPU platforms (DGX/HGX).
• Knowledge of CUDA and GPU workloads.
• Ability to work with engineers, researchers and technical stakeholders.
• Strong problem solving and documentation skills.

Responsibilities

• Administer, maintain, and troubleshoot production HPC environments.
• Manage and support Slurm infrastructure.
• Install, configure, patch, and support RHEL systems.
• Deploy, manage, and troubleshoot Nvidia GPU platforms including DGX/HGX.
• Install, maintain, and upgrade Nvidia drivers and CUDA toolkits.
• Monitor cluster performance and resource utilization.
• Support researchers and engineers running compute workloads.
• Diagnose and resolve hardware, OS, storage, and network issues.
• Collaborate with vendors and stakeholders to improve HPC infrastructure.
• Develop and maintain automation and technical documentation.

Skills

HPC administration

RHEL administration

Slurm workload manager

Nvidia GPU platforms

DGX/HGX systems

CUDA toolkit

GPU drivers

Linux troubleshooting

Tools

Docker

Apptainer/Singularity

Ansible

InfiniBand

Lustre

GPFS

OpenMPI/OpenMP

Job description

LA International seeks an experienced Senior HPC Engineer to support and evolve enterprise HPC platforms used for research, engineering, and computational workloads on a long-term program. The role requires hands-on Linux (RHEL), Slurm, and Nvidia GPU-based environments, including DGX/HGX systems.

The successful candidate will administer, optimize, and troubleshoot production HPC clusters, collaborating with researchers and engineers to deliver infrastructure improvements.