Senior Software Engineer - AI Compute Libraries & Performance

Responsibilities

• Design and implement kernels for linear algebra and tensor ops (GEMM, batched GEMM, convolutions, reductions, elementwise and fused operations) in C++
• Own performance and correctness through microbenchmarks, regression tests, and numerics validation
• Profile and optimise for next-generation AI hardware, focusing on threading, cache locality, memory layout, and kernel launch efficiency
• Debug issues, resolve bugs, and improve product quality and functionality
• Engage in and support Agile ways of working within the team
• Mentor colleagues, sharing knowledge and providing guidance

Requirements

• Excellent programming and scripting skills in C++ and Python
• Understanding of processor architectures and profiling on Linux
• Excellent written and oral communication skills
• Strong teamwork ethic and commitment to quality

Preferred Qualifications

• Strong command of algorithmic performance: vectorisation, memory hierarchy, threading, lock-free patterns
• Hands-on experience with at least one BLAS/DNN stack and ability to read/extend kernels
• Comfort with CPU micro-optimisations and numerical stability across FP32/FP16/BF16/FP8
• Experience integrating native code into PyTorch or similar frameworks
• Knowledge of ABI/API stability and packaging for Linux systems (manylinux, wheels)

Benefits

• Competitive salary
• Annual leave policy
• Medical and dental health plans
• Gym card
• Employee pension matched up to 4%
• Inclusive work environment with equal opportunity process
• Flexible interview approach with reasonable adjustments available

About the Company

Graphcore builds the future of AI compute. As part of the SoftBank Group, the company creates complete AI compute stacks from silicon and software to datacenter-scale infrastructure, delivering key technology into the fast-growing SoftBank AI ecosystem.