HPC Infrastructure Engineer

Job Summary

Joining a small research infrastructure team, the full-time remote HPC Infrastructure Engineer will operate and improve NVIDIA GPU clusters, focusing on automation, performance tuning, and hardware management to enhance research throughput and efficiency.

Key responsibilities

• Operate and enhance the GPU fleet, including provisioning, scheduling, monitoring, and upgrades
• Build automation to maintain fleet health and streamline operations without human intervention
• Evaluate and benchmark rented GPU capacity while ensuring security and access control for clusters

Required qualifications

• Experience managing large-scale Linux server or GPU environments in production
• Proficient with the NVIDIA stack, including drivers, CUDA, and NCCL
• Strong skills in automation using Python and/or Bash, with familiarity in IaC tools like Ansible or Terraform
• Ability to analyze metrics and logs to troubleshoot performance issues effectively
• Comfortable with hands-on hardware work and datacenter operations