4. Job Description Summary

What You'll Do:

• Build optimisers, code generators, runtime libraries and orchestration systems.
• Develop and improve compute pipelines across heterogeneous hardware.
• Investigate hardware behaviour and optimise performance-critical execution paths.
• Work across the stack, from high-level orchestration to low-level hardware performance.
• Explore unfamiliar technical areas and develop practical solutions.
• Collaborate with stakeholders to understand problems and build effective tools.

Your responsibilities will include:

• Designing and evolving machine learning frameworks for research and production.
• Improving the speed and efficiency of research iteration.
• Optimising production workloads to run efficiently on available hardware.
• Working with C++, Rust, Python, MLIR and CUDA.
• Analysing compute graphs and identifying performance bottlenecks.
• Building cohesive systems across the full compute stack.

Why Join Us:

• Work on technically challenging problems spanning software and hardware.
• Make a direct impact on research productivity and production performance.
• Work across a broad technology stack rather than a single speciality.
• Collaborate closely with engineers, researchers and business stakeholders.
• Learn from complex problems in a fast-paced, highly technical environment.

About You:

• You enjoy moving between programming languages and technologies.
• You care about choosing the right tool for the problem.
• You like understanding how software compiles and interacts with hardware.
• You're comfortable learning unfamiliar parts of a technical stack.
• You're naturally curious and enjoy solving open-ended problems.
• You communicate effectively with both technical and non-technical stakeholders.
• You don't need a finance background or experience with every technology. Bring your curiosity, engineering mindset and appetite for difficult problems—we'll help you learn the rest and make an impact from day one.