Machine Learning Performance Engineer

The position involves enhancing the efficiency of machine learning models through detailed systems-level optimisations including GPU, storage, and networking considerations. The engineer will work with modern ML techniques, performance debugging tools, and specialised libraries to optimise training and inference processes. Expertise in low-level GPU programming, CUDA, and distributed training technologies is essential to address latency and throughput challenges. The role requires a comprehensive understanding of hardware architectures, networking technologies, and high-performance computing principles. Analytical and inventive skills are valued to develop and evaluate novel performance strategies. Fluency in English is necessary for collaboration and documentation purposes.