Senior HPC Systems Engineer: GPU Clusters & SLURM

Hardware FYI; Costa Mesa (CA); On-site; USD 147,000 - 196,000; Full time

Job summary

Anduril Industries is seeking an HPC System Engineer to design, deploy, and maintain large-scale HPC infrastructure supporting national security programs. You will work on GPU-enabled clusters, integrate Slurm and cluster software, and ensure performance across global deployments.

You will optimize storage and interconnects, manage AI/ML software stacks (CUDA, PyTorch, TensorFlow), and mentor teammates while ensuring security and compliance.

Qualifications

• 5+ years designing, developing, and implementing large scale compute enterprise systems.
• Strong HPC concepts including cluster architecture, file systems, and high-speed interconnects.
• Experience with Red Hat Enterprise Linux or Ubuntu, HPC, GPU, and cloud services (AWS/Azure).
• Proficiency with systems automation tools (Ansible, Salt, Puppet).
• Experience with HPC file systems (Lustre, GPFS, Pure, VAST) and batch schedulers (SLURM, PBSPro).
• Ability to lift 50 lbs and eligibility to obtain a US Top Secret Clearance.

Responsibilities

• Architect and deploy GPU infrastructure, leading design, deployment, and lifecycle management of NVIDIA hardware (H100/H200, B200/B300).
• Rack, stack, cable, and configure physical servers and multi-node GPU systems end-to-end.
• Configure HPC environments and schedulers (Slurm, multi-user login, cluster management).
• Manage high-speed interconnects (NVLink, NVSwitch, InfiniBand/NDR) for distributed training.
• Manage large-scale storage platforms (petabytes) optimized for AI/ML data access patterns.
• Install and maintain the application stack on HPC clusters (StarCCM+, Ansys, Matlab, CUDA, PyTorch, TensorFlow).
• Implement GPU virtualization (MIG) to maximize resource utilization.

Skills

HPC concepts

System administration

Cloud services

Automation tools

HPC schedulers

GPU hardware

Storage systems

Security clearance

Tools

SLURM

Warewulf

NVIDIA DCGM

CUDA

Job description

Anduril Industries is seeking an HPC System Engineer to design, deploy, and maintain large-scale HPC infrastructure supporting national security programs. You will work on GPU-enabled clusters, integrate Slurm and cluster software, and ensure performance across global deployments.

You will optimize storage and interconnects, manage AI/ML software stacks (CUDA, PyTorch, TensorFlow), and mentor teammates while ensuring security and compliance.