Senior GPU Systems Engineer: Scale AI Clusters & HPC

Career Techniques; New York (NY); Hybrid; USD 200,000 - 300,000; Full time

Job summary

Career Techniques in New York seeks an experienced infrastructure engineer to design, deploy, and scale large-scale GPU clusters for AI research. You will work across compute, storage, OS, and automation to support hundreds of petabytes and thousands of nodes.

You will profile GPU workloads, remove bottlenecks, and collaborate with researchers to translate findings into speedups. Expect to own end-to-end infrastructure projects from design through long-term support and vendor engagement.

Qualifications

• 5+ years engineering large-scale Linux systems in HPC/AI or distributed infra.
• Strong Linux fundamentals, including kernel-level troubleshooting.
• Hands-on debugging of distributed GPU workloads with a strong mental model of GPU performance.
• Experience with GPUDirect RDMA and data movement between GPUs and the network.
• Python for automation; CUDA or C/C++ for reading, profiling, and debugging GPU code.
• Familiarity with configuration tools such as Salt, Ansible, Puppet, or Chef.
• Ability to diagnose cross-stack issues across hardware, OS, and network.
• Clear communication with researchers, engineers, and vendors.

Responsibilities

• Design, deploy, and scale distributed GPU clusters from hardware selection to production operation.
• Profile GPU workloads and identify performance bottlenecks across compute, storage, and network.
• Collaborate with researchers to benchmark workloads and implement speedups.
• Build automation to operate thousands of nodes, including provisioning and self-healing.
• Own infrastructure projects end-to-end from scope to long-term support.
• Engage with vendors to root-cause complex hardware and software issues.

Skills

Linux systems

GPU workloads

Python

CUDA/C++

GPUDirect RDMA

Diagnostics

Communication

Tools

Salt

Ansible

Puppet

Chef

Job description

Career Techniques in New York seeks an experienced infrastructure engineer to design, deploy, and scale large-scale GPU clusters for AI research. You will work across compute, storage, OS, and automation to support hundreds of petabytes and thousands of nodes.

You will profile GPU workloads, remove bottlenecks, and collaborate with researchers to translate findings into speedups. Expect to own end-to-end infrastructure projects from design through long-term support and vendor engagement.