Senior GPU Infrastructure Engineer — InfiniBand / Linux

We are building U.S.-based GPU infrastructure for AI training and inference, with planned deployments involving NVIDIA B300 and RTX PRO 6000 servers.

We are looking for a hands-on Senior GPU Infrastructure Engineer to help design, deploy, validate, and operate our GPU clusters. You will work directly with the founding team, server manufacturers, networking providers, and data-center partners.

What you’ll do

• Review server and networking BOMs, network topology, cabling plans, and hardware compatibility before deployment.
• Deploy and troubleshoot NVIDIA/Mellanox InfiniBand networks and Ethernet networks for management, storage, and customer connectivity.
• Configure Linux servers, GPU drivers, NIC firmware, and supporting infrastructure within the agreed customer service scope.
• Validate GPU health and cluster performance using tools such as DCGM, NCCL tests, and InfiniBand diagnostics.
• Define acceptance criteria, document test results, and support customer handover.
• Establish monitoring, alerting, configuration backups, and operational runbooks.
• Troubleshoot hardware, network, and performance issues with OEM support and data-center remote-hands teams.
• Participate in planned deployments, maintenance, and an agreed on-call support arrangement.

What we’re looking for

• 5+ years of relevant experience in infrastructure, Linux systems, networking, or data-center engineering.
• Hands-on experience deploying or operating multi-node GPU clusters.
• Practical InfiniBand experience, including fabric configuration, subnet management, link diagnostics, and performance troubleshooting.
• Strong Linux administration and Ethernet networking fundamentals.
• Ability to distinguish network, driver, hardware, and workload-related performance issues.
• Experience with scripting or automation, such as Python, Bash, or Ansible.
• Clear documentation and communication skills, with the ability to take ownership of technical delivery.

Experience with H100, H200, A100, or comparable GPU clusters is welcome; prior B300 experience is not required.

Preferred qualifications

• NVIDIA NCP-AIN or NCP-AII certification.
• Cisco CCNP Data Center or CCIE Data Center certification.
• Experience with NVIDIA UFM, ConnectX adapters, RDMA, and GPU cluster monitoring.
• Experience with bare-metal provisioning, customer acceptance testing, or infrastructure operations at a GPU cloud provider.

Certifications are a plus. Demonstrated deployment and troubleshooting experience is our primary selection criterion.

How to apply

Please include your resume and a brief description of a GPU cluster you helped deploy or operate: GPU type, node count, networking technology, your responsibilities, and one technical problem you personally resolved.