Staff System Tests & Diagnostics Engineer

Graphcore is one of the world’s leading innovators in Artificial Intelligence compute. It is developing hardware, software and systems infrastructure that will unlock the next generation of AI breakthroughs and power the widespread adoption of AI solutions across every industry. As part of the SoftBank Group, Graphcore is a member of an elite family of companies responsible for some of the world’s most transformative technologies. Together, they share a bold vision: to enable Artificial Super Intelligence and ensure its benefits are accessible to everyone.

About the Role

System Tests & Diagnostics Engineer – develop, extend, and integrate specialized silicon validation and diagnostics tools for next‑generation AI SoCs. Unlike traditional validation roles focused on executing test plans, this position is responsible for developing the diagnostic software and stress tools that expose hardware failures, characterize silicon behavior, and improve platform observability throughout bring‑up and validation. You will work closely with Arm engineers to understand and extend existing diagnostics technologies while developing Graphcore‑specific capabilities for future AI hardware.

Responsibilities

• Learn and extend existing Arm diagnostics tools to support Graphcore’s AI platform.
• Develop specialized diagnostics and stress‑testing software for CPU, memory, interconnect, PCIe, and other system components.
• Build reusable validation utilities that integrate into the existing validation framework.
• Develop tools that expose intermittent hardware failures including silent data corruption, timing‑related issues, and hardware instability.
• Design configurable diagnostic applications supporting multiple silicon revisions, platforms, and execution environments.
• Develop automation for executing diagnostics across server and rack‑scale validation environments.
• Design configuration‑driven diagnostics supporting scalable parameterization and workload variation.
• Collaborate with hardware architects to understand new hardware capabilities and identify missing diagnostics.
• Work closely with firmware and validation teams to improve hardware observability.
• Debug low‑level hardware, firmware, operating system, and driver interactions.
• Analyze diagnostic output and improve fault isolation methodologies.
• Develop software that enables engineers to reproduce and investigate difficult hardware failures.
• Contribute reusable libraries and utilities that improve engineering productivity across multiple validation teams.

Requirements

• Bachelor’s or Master’s degree in Computer Science, Computer Engineering, Electrical Engineering, or a related technical discipline.
• 10+ years of experience developing hardware diagnostics, system validation software, firmware validation tools, or low‑level systems software.
• Strong software development skills in Python and C/C++.
• Strong Linux systems experience.
• Deep understanding of modern server platform architecture (CPU, memory, PCIe, storage, networking, firmware, OS, device drivers).
• Experience debugging hardware/software interactions across multiple platform layers.
• Experience developing diagnostics, stress testing, or hardware validation software.
• Experience with server platforms, embedded systems, or SoC validation.
• Strong analytical and debugging skills.
• Proven ability to collaborate across hardware, firmware, software, and validation teams.
• Excellent communication and problem‑solving skills.

Preferred Qualifications

• Experience with Arm‑based server platforms.
• Background in AI accelerators or high‑performance computing systems.
• Knowledge of Silent Data Corruption (SDC) testing.
• Experience with hardware stress testing and power transient analysis.
• Familiarity with hardware fault injection methodologies.
• Firmware validation and silicon bring‑up experience.
• Performance characterization and hardware telemetry/observability.
• Experience with BMC firmware or server management RAS technologies.
• Validation across subsystems such as CPU scaling and cache behavior, memory (DDR/HBM) bandwidth, latency, NUMA effects, interconnect contention, PCIe/I‑O throughput, latency, multi‑device scenarios, and high‑speed I/O.

What Success Looks Like

• Extend existing diagnostics technologies to support new Graphcore hardware.
• Develop new diagnostics tools for emerging hardware capabilities.
• Build reusable software components that integrate seamlessly with the existing validation framework.
• Improve hardware observability and root‑cause isolation.
• Detect and characterize intermittent hardware failures that traditional