Senior Applied Power Architect - GPU

NVIDIA NVIDIA · Semiconductors · Santa Clara, CA +1

NVIDIA is seeking a Senior Applied Power Architect to design and optimize power efficiency for next-generation GPUs, focusing on AI architectures. The role involves inventing methodologies, driving power management initiatives, analyzing electrical/thermal constraints, and collaborating with cross-functional teams. Requires MSEE/MSCE or PhD with specialization in low-power-processor architectures and 6+ years of experience.

What you'd actually do

  1. You will be responsible for inventing methodologies, features and techniques to address electrical/speed/power challenges for NVIDIA's next generation AI architectures such as Rubin and Feynman.
  2. Drive power management initiatives to optimize power efficiency and performance, while addressing electrical/thermal and power constraints.
  3. Understand process characteristics and its impact of V-F and drive analysis of various V-F tradeoffs considering Vt choices on power/other metrics and overall performance.
  4. Investigate multi-die technologies and come up with innovative solutions to address packaging, interconnect, physical design, electrical and thermal challenges.
  5. Analyze peak current, Di/Dt, IR and EM requirements, and collaborate closely with ASIC designers, Power Integrity, Packaging experts and Product engineers to understand challenges and formulate solutions.

Skills

Required

  • MSEE/MSCE, preferably PhD, or equivalent experience with a specialization in low-power-processor architectures
  • Solid understanding of energy efficient system design fundamentals, performance/power modeling and related tradeoffs
  • 6+ years of relevant industry experience
  • Strong interpersonal and organizational skills
  • ability & desire to work as a great teammate

Nice to have

  • A background in PDN design and good understanding of electrical issues such as Di/Dt, undershoot/overshoot, IR, EM, peak current, analysis/mitigation techniques
  • Exposure and working knowledge of Python and data analysis packages like: Pandas, NumPy, PyTorch
  • Exposure to lab setup including power measurement equipment such as scope/DAQ
  • ability to analyze board-level power issues like supply voltage, over-current etc.

What the JD emphasized

  • next generation AI architectures
  • power efficiency