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. This role focuses on turning measurements from large-scale distributed systems into engineering decisions that determine production readiness.
You will analyze performance and reliability data collected from Linux clusters (rack scale and larger), working with measurement data from compute, network, and ML workloads. The core challenge is interpreting results correctly—understanding performance variability, defining what "normal" and "acceptable" system behavior looks like, and helping establish pass/fail thresholds.
Typical work includes:
- Analyzing results from automated test frameworks (e.g., pytest-based systems)
- Comparing results across runs, configurations, and system scales
- Working with real-world, imperfect data to support engineering decisions
- Influencing how measurements are designed to produce better data
- Improving how results are stored, queried, and interpreted
This is not a traditional data analysis or BI role. The focus is on understanding system behavior and supporting critical engineering decisions. You'll need to be comfortable reasoning about distributed systems performance, focusing on evidence and correctness rather than presentation.
The role is based in Gdańsk with a hybrid working policy (2-3 days per week in office). Graphcore is expanding teams globally to solve the toughest problems in AI infrastructure, bringing together semiconductor, software, and AI experts to create the complete AI compute stack.
