Staff / Principal Machine Learning Engineer, Serving - UK

📋 Description Build real-time agentic ML serving systems. Optimize inference with vLLM and TRT-LLM. Develop high-performance GPU code: C++, CUDA, Rust. Scale multi-GPU/multi-node inference (Kubernetes, Ray). Own models from research to production. Contribute to open-source projects and docs. 🎯 Requirements Inference optimization with vLLM and TRT-LLM. Model acceleration: quantization, distillation, caching, batching. High-performance code: C++, CUDA, Rust, or optimized Python; GPU profiling. Distributed systems: Kubernetes, Ray, multi-GPU/multi-node inference. Full-cycle ownership: containerize models to production. PhD or equivalent experience in backend/ML systems. 🎁 Benefits Equity and benefits included. Flat structure with fast iterations. Visible work and open-source contributions. Potential relocation support for US move. Work with leading researchers in realtime voice ML. Impactful, production-grade ML systems.