The Compute Node team builds services for managing Virtual Machines running on GPU servers, integrating with disk management, virtual and InfiniBand networks.
The team is also responsible for developing the Virtual Machine Scheduler that operates across clusters with thousands of servers and tens of thousands of GPUs, spanning multiple data centers in several regions.

Experience with Kubernetes not only for deployment but for development is highly valued.
KubeVirt and QEMU/KVM familiarity is a high plus.
Solid knowledge of internal OS architecture, performance considerations, process isolation and resource management.
Knowledge of POSIX, sysfs, system calls, and file systems.
Familiarity with server architecture, PCIe devices, NICs, and kernel drivers.
Experience or strong interest in working with GPUs, DPUs, or ARM architectures is desirable.
Familiarity with the NVIDIA DOCA Software Framework will be an advantage.
Knowledge of at least one of Go or C++, ready to learn or work with the others as needed.
Strong grasp of concurrency, debugging, and profiling techniques.

Competitive salary and comprehensive benefits package.
Opportunities for professional growth within Nebius.
Flexible working arrangements.
A dynamic and collaborative work environment that values initiative and innovation.

System Engineer – Compute Node

Key skills