Advanced Micro Devices logo

Power Methodology Engineer

Advanced Micro Devices
Full-time
Remote friendly (Santa Clara, CA)
Worldwide
Level - Senior

Role Summary

The Power Methodology Engineer will serve as the primary lead for designing power architecture solutions, focusing on microprocessors, GPUs, and machine learning accelerators. The role involves creating power management algorithms and developing power models while working with cross-functional teams to validate power-saving features through the design cycle.

Experience Level

This position requires extensive industry experience, particularly in low-power-processor architectures or power management.

Responsibilities

Key duties include:

  • Optimizing energy efficiency and power delivery for high-performance computing hardware, particularly in AI and machine learning applications.
  • Driving power methodology for AI hardware components and utilizing simulation tools to optimize power consumption.
  • Conducting workload and power analysis for various hardware components like NPUs, GPUs, and CPUs.
  • Maximizing performance within power and thermal limits for data centers and gaming applications.
  • Estimating and analyzing power consumption across different design stages (architecture, RTL, physical design).
  • Creating power models and scripts for performance and power trade-offs.
  • Researching and developing methodologies to enhance power analysis efficiency.
  • Collaborating with teams across multiple domains to ensure power requirements are met.
  • Mentoring junior team members and providing technical leadership on projects.

Requirements

The ideal candidate should possess the following qualifications:

  • Extensive experience in ASIC/SoC power analysis and optimization techniques.
  • Familiarity with machine learning algorithms and their application to power estimation.
  • Proficiency in hardware description languages (Verilog, VHDL) and scripting (Python).
  • Strong analytical and problem-solving skills for complex multidisciplinary challenges.
  • Excellent communication and collaboration skills.

Education Requirements

A Master's degree or PhD in Electrical or Computer Engineering is preferred.