Staff Research Engineer, Model Efficiency

The Staff Research Engineer, Model Efficiency at Cohere is responsible for enhancing the inference efficiency of Large Language Models (LLMs) within AI systems. This role involves developing and deploying novel techniques to optimize model architecture, decoding algorithms, and software/hardware co-design for GPU acceleration. Candidates should possess a PhD in Machine Learning, significant experience in model efficiency techniques, strong software engineering skills, and a background in research publications.