Staff Cloud Hypervisor Engineer, Compute
About The Position
Crusoe is seeking a Staff Cloud Hypervisor R&D Engineer to lead architecture and development of its next-generation virtualization stack. The role focuses on designing a greenfield hypervisor environment in which GPUs, DPUs, and high-speed interconnects are first-class components, reducing virtualization overhead and delivering bare-metal performance for demanding AI and machine-learning workloads. The engineer will lead research and implementation of core hypervisor components, develop hardware pass-through and abstraction approaches for NVIDIA GPUs and BlueField DPUs in multi-tenant environments, and address challenges such as live migration of AI workloads with large VRAM allocations and PCIe peer-to-peer communication between virtualized accelerators. The position also involves profiling and optimizing the full stack, contributing to open-source virtualization projects, and designing security boundaries that balance hardware performance with tenant isolation. This is a technical leadership role requiring collaboration across hardware engineering and cloud control-plane teams. Crusoe offers competitive compensation and equity, along with benefits and programs that may vary by location. The stated compensation range is up to $215,000–$260,000 plus bonus; Restricted Stock Units are included in all offers.
Requirements
- At least 7 years of experience in hypervisor internals, kernel development, or low-level systems programming.
- Expert-level knowledge of CPU virtualization, including Intel VT-x and AMD-V, and memory virtualization, including EPT/NPT and HugePages.
- Mastery of C and C++.
- Experience with specialized AI hardware, including GPUs, InfiniBand/RoCE NICs, and SmartNICs/DPUs.
- Deep understanding of VirtIO, vhost-user, and hardware-accelerated I/O paths.
- Experience leading complex cross-functional projects involving hardware engineering and cloud control planes.
Nice To Haves
- Proficiency in Rust for systems programming.
- Experience with QEMU and KVM.
- Experience with Linux kernel debugging tools such as perf, ftrace, and eBPF.
Responsibilities
- Lead research and implementation of core hypervisor components, including KVM, QEMU, or custom Rust-based solutions, optimized for large GPU fleets.
- Develop and refine hardware pass-through and abstraction techniques, including SR-IOV, VFIO, and mdev, for NVIDIA GPUs and BlueField DPUs in multi-tenant environments.
- Address technical challenges including live migration for AI workloads with 80GB-plus VRAM and PCIe peer-to-peer communication between virtualized accelerators.
- Analyze performance bottlenecks across CPU microarchitecture, MMU virtualization, and guest OS scheduling to reduce jitter and maximize throughput.
- Contribute to and maintain upstream open-source virtualization projects and participate in Linux kernel and virtualization communities.
- Design security boundaries for AI-native cloud infrastructure, balancing high-performance hardware access with multi-tenant isolation and hardening.
- Lead complex cross-functional projects connecting hardware engineering and cloud control planes.
Benefits
- Competitive compensation and equity packages
- Restricted Stock Units
- Paid time off, paid holidays, and leave of absence programs
- Comprehensive health, dental, and vision insurance
- Employer contributions to HSA account
- Paid parental leave
- Paid life insurance
- Short-term and long-term disability
- Professional development and tuition reimbursement
- Mental health and wellness support
- Commuter benefits for parking and transit
- Cell phone stipend
- 401(k) retirement plan with company match up to 4% of salary
- Volunteer time off
- Global travel insurance and emergency assistance
- Daily meals allowance
- Additional perks and programs specific to location
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Senior
Education Level
No Education Listed
