한국어·English
Nebius company · OPEN ROLE
- Company
- Provides GPU cloud infrastructure needed to train AI models.
- Role
- Designs cluster architecture connecting tens of thousands of GPUs.
Log in to mock apply, save this role or see more roles.
What you will do
We rewrote the responsibilities and qualifications in plain language instead of copying the original posting.
Responsibilities
- Architects scalable GPU cluster topologies, including compute nodes, interconnect, storage, and control planes.
- Analyzes AI/ML workloads such as LLM training and inference to inform tradeoffs across latency, bandwidth, and GPU density.
- Validates low-latency, high-throughput interconnects such as InfiniBand HDR/NDR and RoCEv2 at POD and data center scale.
- Works with storage teams to optimize performance for training datasets and checkpointing.
- Analyzes monitoring signals to detect design flaws.
- Partners with site reliability, networking, storage, and data center engineering to operationalize and scale the architecture.
What you need
- Has 5+ years of experience designing clusters.
- Has deep understanding of modern GPU architecture such as NVIDIA and AMD.
- Has experience with HPC interconnects such as InfiniBand and RoCE.
- Has a solid background in systems architecture, networking, and hardware reliability.
- Can script automation and telemetry pipelines in Python, Go, or similar.
Read the employer’s original job description
See how your resume reads for this role
A former Korea GM at a global company reads it and tells you what to improve for this role. It uses 1 credit.
Checked 2026-10-19
NebiusGPU Cluster ArchitectStart a mock application