Staff ML Systems Engineer, Inference

20 hours ago

Winnipeg, Manitoba, Canada Cohere Full-time
Cohere is seeking an experienced engineer to join the Model Efficiency team, focusing on building reliable ML systems and optimizing LLM inference. You will improve the inference stack, reduce latency, and boost throughput across diverse workloads. The role offers remote-friendly work with opportunities across global offices. Ideal candidates have 5+ years of high-performance coding experience in C++ or Python and familiarity with LLM ecosystems and CUDA optimizations.