Requirements

Experience managing and developing engineers
A strong background in computer science, such as BSc, MSc, PhD or equivalent experience
Proven expertise in profiling, benchmarking and optimising distributed or large-scale workloads
Proficiency in one or more programming languages, with strong grounding in algorithms and performance optimisation
Deep understanding of Linux internals, for example across scheduling, memory management, NUMA, networking and filesystems
Experience with HPC schedulers and Kubernetes workload orchestration
Hands-on experience with heterogeneous compute, such as GPUs, multi-core CPU workloads and high-memory systems
Excellent communication and collaboration skills across research, infrastructure and engineering domains
Familiarity with profiling and monitoring tools, including perf, eBPF, VTune, Flamegraphs, Prometheus and Grafana

What you'll be doing

Defining and delivering the long-term strategy for performance engineering across G-Research, enabling all engineers to maximise utilisation of our compute estate
Providing leadership, mentoring and career development for engineers, fostering a high-performance culture
Communicating performance improvements and team impact effectively across technical and non-technical audiences
Engaging directly with researchers, senior stakeholders and engineers to understand compute challenges and design optimised solutions
Profiling, benchmarking and tuning large-scale workloads across CPU, GPU and memory-intensive jobs
Developing reference implementations, libraries and tools to improve job efficiency and reliability
Collaborating with systems, architecture and platform teams to evolve our compute stack
Influencing long-term platform and infrastructure decisions