Machine Learning Performance Engineer

A Machine Learning Performance Engineer with system optimisation skills is needed to enhance model training and inference processes. The role involves analysing hardware and software systems, including GPU architecture, storage, and networking, to improve throughput and latency. Candidates should have experience with low-level GPU programming, debugging tools, and knowledge of relevant libraries and technologies. Expertise in networking technologies linking GPU clusters and understanding distributed training algorithms are also desired. The position demands a curious mindset, problem-solving skills, and proficiency in English. The engineer will work within a dynamic environment where continuous performance improvement is vital for ML applications.