GPU Virtualization

About the Role:

OpsMaven is looking for a highly skilled Senior GPU Virtualization / Live-Migration Engineer to join our team and support a leading US-based client, Elpho. This role is ideal for an engineer with deep expertise in Linux systems, virtualization technologies, and GPU infrastructure who enjoys working close to the operating system, hypervisor, and kernel layers.

The successful candidate will play a key role in developing and optimizing GPU virtualization capabilities, live migration workflows, and cloud-native GPU infrastructure. You will collaborate with architects and engineering teams to build scalable, high-performance platforms that support next-generation AI, HPC, and GPU-intensive workloads.

KEY ROLES & RESPONSIBILITIES:

• Develop and enhance GPU live migration capabilities across virtualized and containerized environments.
• Design, implement, and optimize solutions within the virtualization stack, including: KVM/QEMU, libvirt, KubeVirt
• Build and maintain CRIU-based checkpoint and restore workflows for workload mobility.
• Develop GPU virtualization solutions leveraging: NVIDIA vGPU, VFIO, SR-IOV, NVIDIA MIG
• Troubleshoot and resolve complex kernel-level and virtualization-related issues.
• Collaborate with architects, platform engineers, and infrastructure teams to design and implement scalable GPU solutions.
• Optimize virtualization performance, resource allocation, and reliability for enterprise-scale deployments.
• Participate in code reviews, technical discussions, and architecture implementation planning.
• Contribute to automation, testing, and performance benchmarking initiatives.

Required Qualifications:

• 7-10+ years of experience in Linux systems engineering, platform development, or virtualization engineering.
• Strong expertise in KVM/QEMU internals and virtualization technologies.
• Strong programming experience in C and/or Rust.
• Hands-on experience with Linux kernel development, debugging, and performance tuning.
• Practical experience with one or more of the following: GPU Live Migration, CRIU, NVIDIA vGPU / VFIO, libvirt
• Experience working with Kubernetes and cloud-native virtualization platforms.
• Strong troubleshooting skills across operating systems, hypervisors, and infrastructure components.
• Ability to work effectively in a distributed, collaborative engineering environment.

Preferred Qualifications:

• Experience contributing to open-source virtualization or cloud-native projects.
• Prior experience with organizations such as: NVIDIA, VMware, Red Hat, Major hyperscale cloud providers
• Experience with: RDMA, GPUDirect, PCIe topology optimization, High-performance networking environments
• Familiarity with enterprise-scale GPU cloud infrastructure and AI/HPC workloads.
• Exposure to KubeVirt, container runtimes, and emerging GPU orchestration technologies.