AI Accelerator Hardware Test Engineer - Defect Detection

Fractile; Bristol; On-site; GBP 65,000 - 90,000; Full time

Job summary

Fractile is building silicon, systems and software to redefine AI frontiers. We are seeking an engineer to develop and maintain system-level hardware defect detection tests for an AI accelerator server platform, focusing on fast, robust tests for manufacturing and service environments.

You will translate defect-detection strategy into repeatable test implementations, collaborate with multiple teams, and help identify and address hardware failure modes.

Qualifications

• 3–7 years of experience in system, server, or hardware test engineering, preferably in manufacturing or service environments.
• Solid understanding of server hardware architectures, PCIe fabrics, power delivery, cooling systems, and common failure modes.
• Hands-on experience developing or executing hardware defect detection or diagnostic tests.
• Experience working in Linux-based test environments.
• Familiarity with BMC-based systems and basic use of OpenBMC or Redfish interfaces.
• Strong debug skills and the ability to distinguish test issues from real hardware faults.
• Ability to write clear test code, logs, and documentation for operational use.

Responsibilities

• Develop and maintain system-level hardware defect detection tests for individual servers and rack-based systems.
• Implement tests to detect assembly defects, marginal components, damaged hardware, cabling or interconnect issues, cooling faults, etc.
• Develop service and field diagnostic tests used for RMA triage, repair validation, and preventative maintenance.
• Implement deterministic, fast-running tests with clear pass/fail criteria and actionable failure output suitable for manufacturing and service technicians.
• Develop PCIe defect detection tests to identify lane faults, link degradation, polarity issues, retimer problems, and intermittent connectivity failures.
• Implement stress and soak tests designed to expose latent or marginal hardware issues.
• Debug test failures and hardware issues, performing root-cause analysis and escalating design or process issues as appropriate.
• Analyze factory and field failure data to identify coverage gaps and propose test improvements.
• Support deployment, versioning, and maintenance of test software across factory and service environments.
• Collaborate with software development teams to integrate tests into production system software.
• Collaborate with hardware, manufacturing, service, and reliability teams to ensure tests align with known defect modes and operational needs.

Skills

System hardware testing

Linux test environments

OpenBMC or Redfish basics

PCIe testing

Debug and root-cause analysis

Test code and documentation

Tools

OpenBMC

Redfish

Job description

Fractile is building silicon, systems and software to redefine AI frontiers. We are seeking an engineer to develop and maintain system-level hardware defect detection tests for an AI accelerator server platform, focusing on fast, robust tests for manufacturing and service environments.

You will translate defect-detection strategy into repeatable test implementations, collaborate with multiple teams, and help identify and address hardware failure modes.