Staff Hardware Engineer: AI Compute Systems (Flexible Work)

Graphcore; Austin (TX); On-site; USD 110,000 - 165,000; Full time

Benefits offered by this job

Medical coverage

Dental coverage

Vision coverage

401(k) retirement plan

Commuter benefits

Wellness services

Employee Assistance Program

Job summary

Graphcore is seeking a Staff Hardware Engineer to support Arm-based hardware platforms across lab and data center environments. You will focus on bring-up, validation and troubleshooting for server blades, racks, and rack-scale infrastructure to ensure reliability and performance of next-generation AI systems.

The role requires collaboration with engineering, platform, and data center teams, strong log/telemetry analysis, and hands-on firmware interaction.

Qualifications

• BS in EE/CE/CS or related field.
• Experience with server hardware architectures and board-level debugging.
• Experience analyzing hardware telemetry and power/thermal metrics.
• Hands-on with HPC or AI compute platforms.

Responsibilities

• Lead advanced break-fix troubleshooting for server blades and rack-scale infra.
• Support hardware bring-up and firmware interaction testing.
• Diagnose system-level failures including thermal behavior and BIOS/BMC issues.
• Collaborate with engineering teams for root cause analysis and action plans.
• Develop SOPs and validation docs; mentor junior staff.

Skills

HPC systems

Server hardware

Telemetry analysis

Collaboration

Written and verbal communication

Education

Bachelor’s degree in Electrical Engineering, Computer Engineering, Computer Science, or related discipline

Tools

Python

Bash

Job description

Graphcore is seeking a Staff Hardware Engineer to support Arm-based hardware platforms across lab and data center environments. You will focus on bring-up, validation and troubleshooting for server blades, racks, and rack-scale infrastructure to ensure reliability and performance of next-generation AI systems.

The role requires collaboration with engineering, platform, and data center teams, strong log/telemetry analysis, and hands-on firmware interaction.