ML Performance Engineer: Large-Scale GPU/CPU Optimization

gresearch; Greater London; On-site; GBP 90,000 - 130,000; Full time

Benefits offered by this job

Lunch provided

35 days annual leave

9% pension contributions

Healthcare and life assurance

Cycle-to-work

Monthly company events

Job summary

G-Research in London is seeking an exceptionally skilled ML Performance Engineer to optimise large-scale workloads across our GPU and CPU infrastructure. This hands-on role will profile, benchmark and tune research workloads, delivering reference implementations and tools to improve efficiency and reliability.

You will collaborate with researchers, infrastructure engineers and platform teams to evolve the compute stack, shape architecture and tooling for scalable ML computation.

Qualifications

• Bachelor’s/Master’s/PhD in Computer Science or equivalent.
• Proven track record profiling, benchmarking and optimising distributed workloads.
• Experience with Python.
• Knowledge of CUDA.
• Experience with HPC schedulers and Kubernetes-based workload orchestration.
• Strong understanding of DL frameworks such as PyTorch.
• Strong background in data structures, algorithms and parallel programming on heterogeneous systems.
• Deep understanding of Linux fundamentals (scheduling, memory, NUMA, networking, filesystems).
• Familiarity with profiling/monitoring tools (nsys, ncu, eBPF-based tools).
• Strong communication across research, infrastructure and engineering teams.

Responsibilities

• Collaborating with researchers, senior stakeholders and engineers to understand compute challenges and design optimized solutions.
• Profiling, benchmarking and tuning large-scale training and inference workloads across CPU/GPU/memory systems.
• Developing reference implementations, libraries and tools to improve job efficiency and reliability.
• Collaborating with systems and platform teams to evolve our compute stack.
• Influencing long-term platform and infrastructure decisions.

Skills

Python

PyTorch

Profiling & benchmarking

Distributed computing

Parallel programming

Education

Bachelor’s/Master’s/PhD in Computer Science or equivalent

Tools

CUDA

Kubernetes

nsys

ncu

eBPF tools

Job description

G-Research in London is seeking an exceptionally skilled ML Performance Engineer to optimise large-scale workloads across our GPU and CPU infrastructure. This hands-on role will profile, benchmark and tune research workloads, delivering reference implementations and tools to improve efficiency and reliability.

You will collaborate with researchers, infrastructure engineers and platform teams to evolve the compute stack, shape architecture and tooling for scalable ML computation.