AI Research Engineer

Graphcore Research participates in both fundamental and applied research, to characterise the computational requirements of machine intelligence and to demonstrate how hardware can drive the next generation of innovative AI models. We publish at leading AI/ML conferences (NeurIPS, ICML, ICLR) as well as specialist workshops, and collaborate with other research teams and organisations across the world.

We pride ourselves on being a supportive and collaborative team, where we organise around our individual research interests to solve problems together in domains such as efficient compute, model scaling and distributed training and inference of AI models for multiple modalities and applications, including for sequence- and graph-based data. We’re based across London, Cambridge and Bristol, with projects and discussions that involve all our locations.

Perhaps the best way to get an idea of what we’re all about is to read one of our papers or an article on our blog. If you’re excited to work at the cutting edge of AI supported by new hardware and want to develop your skills in this area, we’d love to hear from you!

Generate AI/ML ideas, design experiments, implement them & evaluate results.

Prepare, submit & present your work to AI conferences and workshops.

Provide technical insight to internal teams by designing experiments and delivering clear, actionable reports.

Collaborate with researchers, silicon and software engineers at Graphcore to help define, build and test Graphcore’s next generation of AI hardware.

Essential:

Master’s, PhD or equivalent experience in a technical discipline (e.g., Maths, Statistics, Computer Science, Physics, Chemistry).

Python programming in a modern deep learning framework, e.g. PyTorch or JAX.

Familiar with deep learning fundamentals: models, optimisation, evaluation and scaling.

Capable of designing, executing and reporting from ML experiments.

Lower-level programming for hardware efficiency, e.g. C++/CUDA/Triton.

Practical familiarity with hardware capabilities for deep learning – threads, caches, vector & matrix engines, data dependencies, bus widths and throttling.

Practical familiarity with software stacks for deep learning – compilation, kernel fusion, XLA/ATen ops, streams, and asynchronous execution

Desirable:

Mathematics skills to support the above: calculus, probability theory and linear algebra.

Experience submitting papers to international scientific conferences or workshops.