Senior Performance Engineer

Performance Engineer – Permanent – London

A specialist research and engineering firm operating at the cutting edge of quantitative computing is investing heavily in its performance engineering capability. The organisation combines deep scientific thinking with large scale infrastructure, creating an environment where complex algorithms meet some of the fastest hardware available. They are seeking an engineer who is driven by understanding how technology behaves at its limits and who wants to turn raw compute power into tangible research progress.

In this role you will work directly with researchers and infrastructure specialists to analyse demanding workloads and remove the barriers that limit speed and efficiency. The position sits at the intersection of software, hardware, and mathematics, requiring curiosity about the whole stack rather than one narrow layer. You will join a team that values careful measurement, practical experimentation, and clear thinking over fashion or buzzwords.

Responsibilities

• Work closely with research teams to understand real computational challenges and translate them into technical improvements
• Analyse and profile training and inference workloads to identify performance constraints
• Design and implement optimisation techniques across operating systems, runtime libraries, and application code
• Create reference tooling and reusable components that help other engineers run jobs more efficiently
• Collaborate with platform and architecture groups to influence the future direction of the compute stack
• Benchmark new hardware and software approaches to guide long term decisions
• Share findings clearly so both technical and non technical audiences can benefit

What you will need

• Academic or professional background in computer science or a related discipline
• Demonstrated experience improving the performance of distributed or large scale workloads
• Strong programming ability with exposure to languages used in systems and machine learning work
• Knowledge of GPU programming concepts and heterogeneous computing
• Solid understanding of Linux fundamentals including scheduling, memory behaviour, and networking
• Experience with HPC style scheduling or Kubernetes based execution
• Familiarity with modern profiling and analysis tools
• Interest in data structures, algorithms, and parallel computation
• Ability to collaborate effectively with researchers and engineers from different disciplines

If you enjoy solving problems where small improvements have enormous impact and want to work close to the heart of a world class research platform, this role offers that opportunity.