Senior Machine Learning Engineer
Summary
Partnering via recruiter Cavendish Professionals, an AI technology company in Germany seeks a Senior ML Engineer to build and scale production model-serving infrastructure. Day to day: optimising inference throughput, latency and GPU utilisation with vLLM/TensorRT-LLM, C++/CUDA/Rust, quantisation and distributed multi-GPU serving on Kubernetes/Ray.
The opportunity
We’re partnering with an AI technology company hiring a Senior Machine Learning Engineer to build, optimise and scale production inference infrastructure.
This role suits an experienced inference or systems engineer who enjoys solving challenges around GPU performance, model serving and distributed systems. The focus is on making models run efficiently and reliably at scale.
What you’ll do
- Build and improve backend infrastructure for serving machine learning models in production.
- Optimise inference performance using vLLM or TensorRT-LLM.
- Improve throughput, latency and GPU utilisation across model-serving workloads.
- Apply model acceleration techniques such as quantisation, speculative decoding, caching, continuous batching and paged attention.
- Scale distributed inference infrastructure across multiple GPUs and nodes.
- Identify performance bottlenecks and improve reliability under high concurrency.
- Make architectural decisions across systems software, model-serving frameworks and infrastructure.
What we’re looking for
- 5+ years of experience as an ML inference or systems engineer, with senior-level technical ownership.
- Hands-on experience with vLLM or TensorRT-LLM for inference optimisation.
- Experience building backend or ML systems and serving models in production.
- Strong systems programming and GPU optimisation skills, using C++, CUDA, Rust or performance-optimised Python.
- Practical experience with model acceleration techniques, such as quantisation, distillation, speculative decoding or attention and batching optimisations.
- Experience scaling distributed systems using technologies such as Kubernetes, Ray, custom load balancing or multi-GPU/multi-node inference.
- Ability to explain performance trade-offs and the impact of your engineering decisions.
Nice to have
- Open-source contributions to inference engines or relevant systems projects.
- Public technical writing demonstrating your expertise.
- Evidence of growing technical responsibility and career progression.
- A PhD in Computer Science, Physics, Mathematics or a related discipline, or equivalent practical systems experience.
- Ability to commute to Mountain View, California.
Skills
Sign in to read and join the discussion.