Hardware Reliability Engineer

Responsibilities

• Plan and execute reliability validation across board, server, and rack levels
• Define and run environmental, accelerated, and mechanical tests including thermal/power cycling, humidity, corrosion, shock & vibration, and HALT/HASS
• Lead shock & vibration validation for transportation, handling, seismic, and operational conditions
• Assess reliability risks for liquid cooling systems such as leakage, fatigue, pump life, corrosion, and coolant stability
• Evaluate HVDC mechanical and electrical robustness for busbars, connectors, and power interfaces
• Perform reliability prediction and life data analysis using Weibull and MTBF
• Lead cross-functional design reviews and drive risk mitigation
• Conduct failure analysis and RCA using standard FA methodologies
• Define and maintain reliability and S&V test specifications (JEDEC, Telcordia GR-63, JESD22, MIL-STD-810, ISTA, ASHRAE, UL, IEC)
• Implement On-going Reliability Test (ORT) for production quality
• Document results and support customer audits and certifications

Requirements

• Bachelor’s or Master’s degree in Mechanical, Electrical, Reliability, Materials, or related Engineering
• 10+ years of reliability engineering experience in AI servers, datacenter systems, HPC, or complex electronics
• Hands-on experience with environmental, shock, and vibration testing
• Strong knowledge of reliability methodologies and statistical analysis
• Practical experience with liquid cooling and HVDC systems
• Proven failure analysis and RCA capability
• Strong communication skills in English; Mandarin a plus

Preferred Qualifications

• AI server architecture and large-scale liquid cooling systems
• FEA/modal analysis and test correlation
• Datacenter, telecom, and transportation standards knowledge
• Reliability certification (e.g., ASQ CRE)

About the Company

At Graphcore, we’re building the future of AI compute. We’re a team of semiconductor, software and AI experts, with deep experience in creating the complete AI compute stack - from silicon and software to infrastructure at datacenter scale.