Director, Sustaining Engineering

Graphcore•Milpitas, CA

4h•Hybrid

About The Position

Graphcore is seeking a Principal Engineer, Power Engineering to help define the architecture of power delivery components from the grid to the chip. The role works with hardware engineers, firmware developers, data center operations, vendors, and other teams to ensure infrastructure meets customer requirements and standards for performance, reliability, and scalability. The posting describes the role as hybrid with a minimum of three days onsite in Milpitas, California, while the overview also says the position will be based in Austin or San Jose and could potentially be onsite or remote within the United States. The position provides technical leadership across data center, rack-level, board-level, and system-level power delivery. It includes establishing a power roadmap, developing and validating hardware and firmware concepts, resolving complex issues, guiding suppliers and engineering teams on testing, and ensuring compliance with industry standards and regulations.

Requirements

Bachelor’s, master’s, or PhD degree in Electrical Engineering with specialized knowledge of Power Engineering.
Demonstrated ability in power system and delivery architecture, design, and development.
Experience with rack-level hardware design, including servers, storage, networking, and power distribution.
Proficiency in power hardware and firmware development and debugging.
Excellent problem-solving and analytical skills.
Strong communication and teamwork skills.
Ability to work in a fast-paced, multifaceted environment.

Nice To Haves

Familiarity with new technologies in AI and data center infrastructure.
Comfort meeting and engaging directly with internal and external customers during requirements gathering and solution development.

Responsibilities

Provide thought leadership, planning, and design for power delivery from the data center power drop through delivery to systems and chips.
Establish the power roadmap and coordinate with vendors to align future technologies and system engineering needs; improve component reliability and efficiency over time.
Specify rack-level power shelf hardware and firmware and design for resiliency.
Set data center power policies for power buffering and oversubscription, job scheduling based on power, and monitoring or accommodating power-quality faults; resolve telemetry and data needs for policy algorithms.
Ensure IT power is robust in the presence of grid disturbances and compatible with grid power requirements.
Apply current board- and system-level power regulation technologies, including VRM, TLDR, and POL, and help develop more efficient technologies while considering manufacturing capabilities and limitations.
Guide power equipment manufacturers, ODMs, and engineering teams on testing and validation procedures for power hardware and firmware.
Develop and test hardware prototypes to explore and validate power-delivery concepts.
Resolve sophisticated power hardware and firmware issues as needed.
Ensure compliance with industry standards and regulations.