ML Infrastructure Engineer

Nebius is looking for an ML Infrastructure Engineer to join their remote team, focusing on benchmarking GPU platforms for machine learning and AI applications. The role involves evaluating GPU performance across various architectures and software stacks, including CUDA and ROCm. Daily tasks will include profiling and analyzing GPU performance at both the system and kernel levels, debugging and optimizing machine learning workloads, and conducting acceptance testing for new GPU clusters to ensure they meet performance and compatibility standards. The engineer will also perform experiments to assess the impact of different system configurations on performance and scalability, as well as develop tools and dashboards for visualizing performance metrics. While the advert does not specify required qualifications or experience, the position is likely suited for individuals with a strong background in hardware and software performance analysis. The role is fully remote, making it accessible to candidates from various locations. This opportunity is ideal for those who thrive in a technical environment and are passionate about advancing AI infrastructure.