Senior HPC Engineer - Systems Integrator

Hamilton Barnes; Texas; On-site; USD 203,000 - 248,000; Full time

Benefits offered by this job

Performance based bonus

Equity incentive

Job summary

Hamilton Barnes in Dallas–Fort Worth is seeking a Senior HPC Engineer to build, optimize and operate bare metal GPU clusters powering AI training and inference. You will work across kernel, GPU, high‑speed fabric and file‑system layers, shepherding HGX clusters from deployment to customer handover.

The role demands strong Linux engineering, InfiniBand/RoCE expertise, Lustre/WEKA experience, and robust Python/Bash scripting.

Qualifications

• Hands-on HPC engineering on bare metal clusters for AI training or scientific computing.
• Strong Linux systems engineering, including kernel, drivers and the RDMA stack.
• Deep InfiniBand experience plus working RoCE experience.
• NVIDIA GPU HGX system experience (CUDA, NCCL).

Responsibilities

• Build, image and burn in NVIDIA HGX clusters, and own fleet baselines, upgrades and rollback.
• Operate and troubleshoot InfiniBand (UFM, topology, routing) and tune RoCE lossless networking.
• Operate and tune Lustre and WEKA, finding the layer at fault in throughput and latency issues.
• Diagnose XID, ECC and NVLink faults, drive vendor cases to closure, and own cluster performance baselines.
• Take HPC escalations from Level 2 and Hardware SRE, and leave a runbook behind.
• Write tested Python and Bash tooling in GitLab, with Ansible as the source of truth.

Skills

HPC engineering

Linux kernel & RDMA

NVIDIA HGX / CUDA

Python & Bash scripting

Tools

Ansible

GitLab

Job description

Keen to join a company that champions growth and development?

Join a fast-growing,vertically integrated innovator in artificial intelligence and high-performance computing (AI HPC) infrastructure.Powered by renewable energy,the organization designs,deploys,and operates large-scale data centers and high-density network fabrics engineered to support world-class AI workloads and massive GPU clusters across North America and global markets.

The company is seeking a Senior HPC Engineer in Dallas Fort Worth to build,optimize,and operate the bare metal GPU clusters powering large-scale AI training and inference.Working directly at the kernel,GPU,high-speed fabric,and file system layers,the successful candidate will take NVIDIA HGX clusters from initial bare metal deployment through to customer handover,owning end-to-end performance and reliability at frontier scale.

Ready to take your expertise to the next level? Get in touch today!

Responsibilities:

• Build, image and burn in NVIDIA HGX clusters, and own fleet baselines, upgrades and rollback.
• Operate and troubleshoot InfiniBand (UFM, topology, routing) and tune RoCE lossless networking.
• Operate and tune Lustre and WEKA, finding the layer at fault in throughput and latency issues.
• Diagnose XID, ECC and NVLink faults, drive vendor cases to closure, and own cluster performance baselines.
• Take HPC escalations from Level 2 and Hardware SRE, and leave a runbook behind.
• Write tested Python and Bash tooling in GitLab, with Ansible as the source of truth.

Skills/Must have:

• Several years of hands on HPC engineering on bare metal clusters for AI training or scientific computing.
• Strong Linux systems engineering, including kernel, drivers and the RDMA stack.
• Deep InfiniBand experience plus working RoCE experience.
• Lustre, WEKA or a comparable parallel file system.
• System level NVIDIA GPU experience on HGX class servers (CUDA, NCCL, firmware).
• Strong Python and Bash, with a base in Dallas Fort Worth (onsite, hybrid or remote).

Benefits:

• Performance based bonus
• Equity incentive

Salary:

• $ 225,000 Base Salary