Senior Machine Learning Engineer (Large Systems)

Graphcore is a company building the future of AI compute, combining semiconductor, software, and AI expertise to create a complete AI compute stack from silicon to datacenter-scale infrastructure. As part of the SoftBank Group, it delivers technology into the SoftBank AI ecosystem and is expanding its teams worldwide.

• Bachelor/Master's/PhD or equivalent experience in Machine Learning, Computer Science, Maths, Data Science, or related field.
• Proficiency in deep learning frameworks like PyTorch and JAX.
• Strong software development skills in Python or C++.
• Expertise in deep learning including training, optimization, and evaluation.
• Experience with distributed training or inference across 64+ accelerators.
• Ability to design, execute, and report on ML experiments.
• Deep understanding of performance bottlenecks and solutions.
• Ability to work quickly in a dynamic environment and collaborate cross-functionally.
• Strong communication skills for explaining technical concepts.

Desirable:

• Experience with MLOps for Kubernetes-based clusters.
• Building production systems with large language models.
• Efficient computing using low-precision arithmetic.
• Writing C++, Triton, CUDA kernels for performance optimization.
• Familiarity with HPC systems and networking technologies like Infiniband, NVLink, and RoCE.
• Contributions to open-source projects or published relevant research.
• Knowledge of cloud computing platforms.
• Willingness to present, publish, and deliver talks in the AI community.

• Implement latest machine learning models and optimize them for performance and accuracy, scaling to thousands of accelerators.
• Test and evaluate new internal software releases, provide feedback, fix code, and conduct code reviews.
• Benchmark models and ML techniques to identify bottlenecks and improve efficiency.
• Design, conduct experiments on novel AI methods, implement and evaluate results.
• Collaborate with Research, Software, and Product teams to build and test next-generation AI hardware.
• Engage with AI community and stay updated on latest AI developments.