AI Kernel & GPU HPC Cluster Engineer

Blue Signal Search; Santa Clara (CA); On-site; USD 150,000 - 210,000; Full time

Job summary

Blue Signal Search seeks an experienced AI Kernel / Cluster Engineer to optimize production GPU and HPC clusters, focusing on Linux kernel level operations and driver management for reliable AI workloads.

You will collaborate with AI and infrastructure teams to ensure high availability, performance, and scalability using SLURM/Kubernetes, NVIDIA tools, and robust automation with Bash and Python.

Qualifications

• Minimum 3 years managing production GPU or HPC clusters.
• Advanced Ubuntu Linux administration experience with kernel-level expertise (driver management, tuning, boot config).
• Production LXC container administration experience.
• GPU networking expertise including InfiniBand and RoCEv2.
• NVIDIA tools (DCGM, UFM, CUDA, NCCL, SHARP) proficiency.
• Experience with SLURM or Kubernetes for workload orchestration.
• Bash/Python automation for diagnostics and operations.

Responsibilities

• Manage and optimize production GPU/HPC clusters for AI workloads.
• Administer Ubuntu Linux at the kernel level, including drivers and tuning.
• Monitor GPU infrastructure health using NVIDIA tools and resolve issues.
• Configure and maintain LXC containers for reliable isolation and networking.
• Support high-performance cluster networking (InfiniBand, RoCEv2).
• Manage workloads with SLURM or Kubernetes to maximize efficiency.
• Develop Bash/Python automation to streamline operations.
• Troubleshoot Linux, GPU, containers, storage, and networking; document solutions.

Skills

Linux troubleshooting

Analytical thinking

Strong communication

Automation scripting (Bash/Python)

Tools

NVIDIA DCGM

NVIDIA UFM

CUDA

NCCL

SHARP

SLURM

Kubernetes

InfiniBand

RoCEv2

LXC

Ubuntu kernel tuning

Job description

Blue Signal Search seeks an experienced AI Kernel / Cluster Engineer to optimize production GPU and HPC clusters, focusing on Linux kernel level operations and driver management for reliable AI workloads.

You will collaborate with AI and infrastructure teams to ensure high availability, performance, and scalability using SLURM/Kubernetes, NVIDIA tools, and robust automation with Bash and Python.