SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Graphcore is building the future of AI compute infrastructure. As part of the SoftBank Group, the company develops semiconductor, software, and AI expertise across the complete AI compute stack—from silicon and software to datacenter-scale infrastructure.
The Performance & Reliability team is responsible for determining whether entire racks of machines meet production standards. This involves measuring and evaluating large-scale Linux systems from single racks to full datacenters, ensuring systems behave correctly and are reliable enough for deployment.
In this role, you'll work across the full spectrum of performance and reliability engineering: designing workloads that expose system behavior, building execution systems to run experiments across large clusters, interpreting results, and defining production readiness criteria. The work combines systems engineering, measurement, and engineering judgment—not pure infrastructure or data work.
You won't be confined to a narrow specialty. Some engineers focus on infrastructure, others on workloads or analysis, but the team collectively owns understanding system behavior at scale. You're expected to specialize in one area while contributing across boundaries.
Essential qualifications include strong software engineering experience across multiple projects or systems over several years, hands-on experience with Linux-based environments (ideally distributed or high-performance systems), Python proficiency, and experience with automation and CI/CD systems. The ideal candidate brings significant practical experience and sound engineering judgment, comfortable working where the right answer isn't obvious and where careful measurement matters more than output volume.
