Performance Engineering Manager - Permanent - London
A technology and research focused organisation operating at the forefront of quantitative innovation is strengthening its performance engineering function. The company runs one of the most demanding compute environments in Europe, where large scale scientific workloads meet modern HPC and Kubernetes platforms. They are seeking a leader who is motivated by deep technical problems and the challenge of turning raw compute power into measurable research advantage. The successful candidate will be expected to be hands on initially, working directly with code, systems, and engineers while the team and capability continue to grow.
In this role you will lead a specialist group responsible for profiling, analysing, and optimising demanding workloads across CPU, GPU, and memory intensive platforms. You will work closely with researchers, infrastructure engineers, and platform teams to remove bottlenecks and design practical solutions that scale with future growth. You will be joining a highly technical environment that values evidence, precision, and thoughtful engineering.
Responsibilities
- Lead and develop a team focused on improving utilisation and efficiency across large compute platforms
- Contribute hands on to profiling and optimisation work, particularly during the early stages of the team
- Profile and benchmark real workloads to understand performance limits and opportunities
- Design and implement tuning strategies across operating system, scheduler, and application layers
- Build reference tools and libraries that help engineers run jobs more reliably and at lower cost
- Partner with research and infrastructure groups to shape future platform design
- Communicate technical improvements clearly to both specialist and non specialist audiences
- Establish long term approaches to monitoring, measurement, and continuous optimisation
What you will need
- Experience managing engineers or computer scientists in a technical environment
- Strong background in computer science and performance focused engineering
- Proven ability to analyse and optimise distributed or large scale workloads
- Solid programming skills with understanding of algorithms and efficiency trade offs
- Deep knowledge of Linux internals including scheduling, memory behaviour, and networking
- Exposure to HPC scheduling and Kubernetes orchestration
- Familiarity with modern profiling and observability tooling
- Practical experience working with GPUs and high memory systems
- A mindset of ownership, collaboration, and long term improvement
If you enjoy working close to the core of a highly technical organisation and want responsibility for performance that directly shapes research output, this role offers that opportunity.
