Lead GPU HPC Infrastructure Engineer

Sciforium; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Benefits offered by this job

Medical, dental, and vision insurance

401k plan

Daily lunch, snacks, and beverages

Flexible time off

Competitive salary and equity

Job summary

Sciforium is seeking a Senior HPC & GPU Infrastructure Engineer to own the health, reliability, and performance of a high-density GPU compute cluster. You will bridge hardware operations, distributed systems, and ML workflows, from Linux systems engineering to CUDA/ROCm stacks and vLLM debugging.

Ideal candidates have 5+ years in HPC/GPU environments, with deep Linux internals knowledge and strong scripting (Bash/Python).

Qualifications

• Requires 5+ years in HPC, GPU cluster operations or similar roles.
• Bachelor’s or Master’s in a technical field.
• Strong CUDA/ROCm and GPU driver debugging experience.

Responsibilities

• Own health, reliability, and performance of the GPU compute cluster.
• Lead deployment of new GPU nodes and topology validation.
• Maintain ML software stack (CUDA, PyTorch, JAX, vLLM) and driver stacks.

Skills

Linux systems engineering

SRE / reliability engineering

Kernel debugging

Networking security

Automation scripting

Python scripting

Education

Bachelor’s or Master’s degree in CS/CE/EE or related field

Tools

NVIDIA GPUs (H100/B200)

AMD GPUs (MI325x/MI355x)

CUDA Toolkit / cuDNN / NCCL

ROCm / ROCm stack

Linux kernel modules

GPFS / Lustre / NFS

NDMA / RDMA networking

SSH / VPN / iptables

Job description

Sciforium is seeking a Senior HPC & GPU Infrastructure Engineer to own the health, reliability, and performance of a high-density GPU compute cluster. You will bridge hardware operations, distributed systems, and ML workflows, from Linux systems engineering to CUDA/ROCm stacks and vLLM debugging.

Ideal candidates have 5+ years in HPC/GPU environments, with deep Linux internals knowledge and strong scripting (Bash/Python).