Senior Staff Data Center Operations Engineer GPU Hardware Architecture
Crusoe is an AI infrastructure company that designs, builds, and operates AI data centers and a cloud platform. It provides managed AI services, GPU compute, model fine-tuning and inference, and infrastructure operations for organizations building and deploying AI workloads.
Maintainer signals as of 8/14/2026
Funding history
Projects
About Crusoe
Crusoe, the AI factory company, provides Crusoe Cloud and Crusoe Intelligence Foundry for AI development and production. Its offerings include managed inference, serverless fine-tuning, high-performance NVIDIA and AMD compute, accelerated storage, RDMA networking, managed Kubernetes and Slurm, and operations tooling. The company also designs, builds, and operates modular AI data-center infrastructure using an energy-first approach, serving customers that need scalable training, inference, and AI platform infrastructure.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will serve as the technical authority for GPU platforms, translate GPU power and thermal roadmaps into facility requirements, analyze fleet telemetry for predictive maintenance, architect spare strategies, create repair SOPs and diagnostic tools, lead complex hardware root-cause analyses, and guide vendor technical relationships.
Requirements
- 10+ years in hardware engineering, systems architecture, or data center infrastructure
- Experience educating and influencing engineering and operations teams
- Experience managing or architecting GPU clusters at scale
- Expert knowledge of NVIDIA Hopper, Blackwell, and Rubin architectures
- Expert knowledge of AMD Instinct architectures
- Mastery of NVLink, NVSwitch, and InfiniBand
- Ability to translate GPU data sheets into mechanical engineering requirements
- Proficiency in Python, Go, or Bash
- Experience with DCGM and ROCm
- Experience using large datasets or machine learning frameworks
- Experience using failure telemetry for sparing and field-service workflows
- Deep understanding of Direct-to-Chip cooling, fluid dynamics, pressure-drop curves, and dripless couplings
- B.S. or M.S. in electrical engineering, computer engineering, or a related technical field
Responsibilities
- Provide technical guidance on upcoming GPU platforms and facility designs
- Translate GPU power and thermal roadmaps into power, cooling, and rack-spacing requirements
- Analyze fleet telemetry for predictive maintenance
- Identify pre-failure patterns in HBM and NVLink components
- Architect site-level spare strategies and define critical spare inventories
- Create SOPs for GPU repairs and manifold maintenance
- Develop diagnostic tools for NVLink, PCIe, and thermal issues
- Act as the Tier-3 escalation point for complex hardware failures
- Lead root-cause analyses across hardware and facility systems
- Maintain a 24-month NVIDIA and AMD architecture roadmap
- Educate stakeholders on HBM, interconnect, and liquid-cooling impacts
- Support OEM and VAR technical relationships
Benefits
- Restricted stock units
- Paid time off
- Paid holidays
- Health insurance
- Dental insurance
- Vision insurance
- Employer HSA contributions
- Paid parental leave
- Paid life insurance
- Short-term disability insurance
- Long-term disability insurance
- Professional development
- Tuition reimbursement
- Mental health support
- Wellness support
- Commuter benefits
- Cell phone stipend
- 401(k) retirement plan
- 401(k) company match up to 4% of salary
- Volunteer time off