Senior Staff Data Center Operations Engineer, GPU Hardware Architecture

Responsibilities

• Provide technical guidance to Data Center Engineering on upcoming silicon architectures like NVIDIA Blackwell/Rubin and AMD MI350/400
• Leverage AI/ML methodologies to analyze fleet-wide telemetry for predictive maintenance
• Architect site-level sparing strategies using failure telemetry and MTBF data
• Create precision SOPs for high-stakes GPU repairs and develop diagnostic tooling
• Act as Tier-3 escalation point for complex hardware failures and lead Root Cause Analysis (RCA)
• Maintain a forward-looking view of NVIDIA and AMD architectures to educate internal stakeholders
• Support technical relationships with OEMs and VARs by auditing hardware builds

Requirements

• Expert-level knowledge of NVIDIA (Hopper/Blackwell/Rubin) and AMD (Instinct) architectures
• Mastery of NVLink, NVSwitch, and InfiniBand physical and logical layers
• Proficiency in Python, Go, or Bash for building telemetry and health-check tools
• Experience with DCGM and ROCm
• Deep understanding of Direct-to-Chip (D2C) cooling and fluid dynamics
• Experience using failure telemetry to inform sparing requirements and workflows

Preferred Qualifications

• 10+ years in Hardware Engineering, Systems Architecture, or Data Center Infrastructure
• Experience managing or architecting GPU clusters at scale (thousands of nodes)
• B.S. or M.S. in Electrical Engineering, Computer Engineering, or a related technical field

About the Company

Crusoe is a vertically integrated AI infrastructure company building climate-aligned AI infrastructure by bridging the gap between energy and intelligence.