SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: PLN 350,700 - 474,400 / annual
Graphcore is building the future of AI compute as part of the SoftBank Group. The company develops the complete AI compute stack—from silicon and software to datacenter-scale infrastructure—and is expanding teams globally to meet growing AI demand.
As a Staff Software Engineer on the ML Kernels & Runtime team, you will design and implement high-performance compute libraries for next-generation AI hardware. Your work will focus on linear algebra and tensor operations (GEMM, batched GEMM, convolutions, reductions, elementwise and fused operations), implemented in C++. You'll own both performance and correctness, adding microbenchmarks, regression tests, and numerics validation to ensure reliability.
Key responsibilities include:
- Design and implement kernels for linear algebra and tensor operations in C++
- Provide technical leadership within the team, helping define approaches, review designs, and guide complex engineering decisions
- Own performance and correctness through comprehensive testing and validation
- Profile and optimize across next-generation AI hardware, focusing on threading, cache locality, memory layout, and kernel launch efficiency
This role is ideal for a senior engineer with deep expertise in low-level performance optimization, numerical computing, and systems programming. You'll work alongside semiconductor, software, and AI experts to solve some of the toughest problems in AI infrastructure.
