Senior HPC Cluster Engineer

About the company

• Nebius is headquartered in the Netherlands, with hubs in Finland, Serbia, and Israel.
• Our own data center in Finland features server racks designed in-house for ML-specific high load, with power-efficient solutions, including a free-cooling system.
• Our mature team of engineers has a proven track record in developing sophisticated cloud and ML solutions and designing cutting-edge hardware.

About the team

The Hypervisor team supports and develops the parts of the Cloud platform that directly affect the KVM hypervisor and QEMU device emulator. We understand the granular details of hardware virtualization and device emulation, paying close attention to performance and protection against untrusted code.

In this position, your responsibility will be to

• Improve infrastructure around GPU-accelerated computing.
• Analyze root cause and suggest corrective action for problems large and small scales.
• Add new hardware support through all infrastructure software stack.
• Detect and fix problems before they occur.

We expect you to have

• 5+ years of professional software development experience.
• 3+ years of experience with Linux.
• Fluency in Go programming language.
• General understanding of QEMU/KVM virtualization stack.

It would be an added bonus if you had::

• System level understanding of server architecture, PCIe devices, NICs, Linux OS and Kernel drivers.
• Experience analyzing and tuning performance for a variety of HPC workloads.
• Familiarity with RDMA, RoCE, InfiniBand.
• Background with Software Defined Networking and HPC cluster networking.
• Familiarity with deep learning frameworks like PyTorch and TensorFlow.

Does this sound like the challenge you've been looking for? If so, we invite you to join us!