Senior ML Performance Engineer – GPU & Systems

Wayve; Greater London; Hybrid; GBP 120,000 - 180,000; Full time

Benefits offered by this job

Equity

Relocation support

Hybrid working

Learning & development budget

Health insurance

Job summary

Wayve is building the leading AI platform for autonomous driving and invites experienced engineers to optimise training and cloud inference workloads. You will profile workloads, identify bottlenecks and implement cross-target optimisations across GPUs and servers.

You will work with Research, model and platform teams to drive performance-driven development, benchmarking and GPU strategy in a hybrid London-based or global environment.

Qualifications

• 10 years of industry experience driving performance engineering across ML systems, GPU compute infrastructure or similar.
• Experience optimizing large-scale GPU workloads for training and inference.
• Proven ability to report and track performance benchmarks openly.
• Strong Python coding, testing and code quality practices.
• BS or MS in ML/CS/Engineering or equivalent experience.

Responsibilities

• Profile ML workloads to identify bottlenecks across training and cloud inference.
• Implement efficiency improvements to maximise MFU, throughput and utilisation.
• Build reusable, cross-target optimisations (kernels, data loaders, frameworks like Triton).
• Develop benchmarking tools to track gains and detect regressions.
• Collaborate with Research, model teams and platform teams on GPUs strategy.

Skills

Performance engineering

GPU compute

Python

Education

BS or MS in ML/CS/Engineering

Tools

Nsight Systems

CUDA

Triton

Job description

Wayve is building the leading AI platform for autonomous driving and invites experienced engineers to optimise training and cloud inference workloads. You will profile workloads, identify bottlenecks and implement cross-target optimisations across GPUs and servers.

You will work with Research, model and platform teams to drive performance-driven development, benchmarking and GPU strategy in a hybrid London-based or global environment.