Senior Systems Engineer – Performance & Reliability

Graphcore is a company building the future of AI compute, comprising a team of semiconductor, software, and AI experts. They create the complete AI compute stack from silicon and software to infrastructure at datacenter scale. As part of the SoftBank Group, they deliver key technology into the SoftBank AI ecosystem and are expanding teams globally to address AI opportunities.

Essential:

• Strong software engineering experience across multiple projects or systems over several years
• Experience working in Linux-based environments, ideally with distributed or high-performance systems
• Proficiency in Python
• Experience with automation and CI/CD systems (e.g., GitLab CI, Jenkins, GitHub Actions)
• Ability to design, implement, and run experiments or tests that produce meaningful results
• Ability to interpret results and communicate findings clearly, emphasizing accuracy and usefulness to decision-making
• Comfortable working in areas where requirements are not fully defined and judgement is required

Desirable:

• Experience working with large-scale or distributed systems (e.g., clusters, cloud platforms, HPC environments)
• Experience with performance, reliability, or systems-level testing/measurement
• Familiarity with pytest or similar frameworks for structured test/measurement execution
• Experience analysing system behaviour under load (compute, network, or ML workloads)
• Experience working with containerisation, orchestration, or provisioning systems (e.g., Docker, Kubernetes, OpenStack)
• Proficiency in other applications programming languages (e.g., C++)
• Exposure to data analysis, statistics, or interpreting variability in results

• Design and run measurements on distributed systems
• Analyse performance and reliability results
• Improve systems that execute measurements at scale

Typical work includes:

• Running measurements on large-scale Linux clusters (rack-level and beyond)
• Using and extending tools such as pytest for measurement execution
• Measuring compute, network, and ML workload performance
• Analysing variability and repeatability of results

Applicants must hold the right to work in Poland. Visa sponsorship or support for visa applications is not provided.