GPU ML Benchmarking Engineer

Nebius Group; Amsterdam; Hybrid; EUR 75,000 - 110,000; Full time

Benefits offered by this job

Competitive compensation

Career growth and learning

Flexibility and ownership

Collaborative culture

Impactful AI projects

Job summary

Nebius is seeking a highly skilled ML/AI Engineer to lead benchmarking of GPU platforms for machine learning and AI workloads. You will evaluate performance across architectures and software stacks, profile at system and kernel levels, and develop tooling to drive platform optimization.

You will work closely with hardware and development teams to debug bottlenecks, perform experiments, and ensure scalable performance for next-generation AI infrastructure.

Qualifications

• Strong ML theory foundations.
• Deep experience with training and inference of large neural networks.
• Experience with PyTorch/TensorFlow/JAX.
• Knowledge of CUDA and GPU driver stack.
• Familiarity with Docker and Kubernetes.
• Excellent communication and ability to work independently.

Responsibilities

• Profile and analyze GPU performance at the system and kernel level.
• Evaluate and compare GPU performance across platforms, architectures, and software stacks (CUDA, ROCm).
• Debug and optimise ML workloads to run efficiently on GPU hardware, identifying bottlenecks.
• Perform acceptance testing for new GPU clusters, ensuring performance, stability and compatibility for AI workloads.
• Develop tools and dashboards to visualise performance metrics, bottlenecks, and trends.
• Contribute to internal tooling, frameworks, and best practices.

Skills

ML theory

PyTorch

GPU profiling

Python

Communication

Tools

Docker

Kubernetes

CUDA

Nsight

nvprof

Job description

Nebius is seeking a highly skilled ML/AI Engineer to lead benchmarking of GPU platforms for machine learning and AI workloads. You will evaluate performance across architectures and software stacks, profile at system and kernel levels, and develop tooling to drive platform optimization.

You will work closely with hardware and development teams to debug bottlenecks, perform experiments, and ensure scalable performance for next-generation AI infrastructure.