Machine Learning Engineer  UK

A Machine Learning Engineer in the UK develops and optimises real-time voice models and inference systems, working with frameworks like vLLM and TRT-LLM. The role involves improving model acceleration through quantisation, distillation, and caching, alongside creating highly performant systems using C++, CUDA, or Python. Responsibilities include managing distributed systems with Kubernetes and Ray, ensuring scalable inference across multiple GPUs and nodes, and contributing to open-source projects or technical documentation. Candidates should have experience in backend or ML systems, preferably with a PhD or equivalent practical expertise, and be comfortable taking models through the full deployment cycle. The role demands a proactive approach to problem-solving, a focus on impact, and an enthusiasm for deep technical understanding, in a fast-paced, innovative environment.