Technical Program Manager - Compute Systems Engineering
About the role
About the team
The Compute Systems Engineering organization owns the full systems stack behind Nebius HPC and GPU Cloud deployments. We develop and operate the core components of our virtualization platform— including the Linux kernel, KVM, QEMU, device drivers, and firmware — together with ultra-high-bandwidth, HPC-grade networking, state-of-the-art NVIDIA GPUs, and other cutting-edge hardware.
In less than two years, we have brought up and successfully operated large-scale B300 and H200 GPU clusters for Cloud customers. These deployments comprise tens of thousands of GPUs across multiple data centers in US, UK, France, and other European countries – with hundreds of thousands more GPUs expected to follow. One of our most significant challenges is continuing to evolve and scale the Nebius Compute platform to support NVIDIA’s flagship GB300 systems and upcoming VR200 solutions.
Our organization is responsible for:
- Regularly adapting our systems stack to support the latest hardware platforms and technologies.
- Developing and maintaining performance improvements and customer-facing capabilities – based on both business requirements and engineering needs.
- Building and operating reliable, comprehensive deployment automation and validation processes.