Senior Compute Kernel Architect, Gpu Power

NVIDIA

CA, United States
Base: 184,000 usd - 287,500 usd (level 4); 224,000...
Cuda kernel development and optimization
Gpu power consumption analysis
Power delivery network (pdn) understanding
Design and develop CUDA kernels purpose-built to enhance GPU power consumption, targeting worst-case current draw across compute, memory, and I/O subsystems

Job Summary

  • Design and develop CUDA kernels purpose-built to enhance GPU power consumption, targeting worst-case current draw across compute, memory, and I/O subsystems.
  • Collaborate with hardware power architects to validate PDN assumptions and di/dt specs, helping ensure chips survive harsh real-world scenarios.
  • This position is an opportunity to have real impact on the GPU power architecture of future NVIDIA products, working at the boundary of GPU architecture, software and silicon.

Matching Summary

Design and develop CUDA kernels purpose-built to enhance GPU power consumption, targeting worst-case current draw across compute, memory, and I/O subsystems.

Salary

Base: 184,000 USD - 287,500 USD (Level 4); 224,000 USD - 356,500 USD (Level 5); Equity and benefits: eligible

Skills & Requirements

Must-have

  • CUDA kernel development and optimization
  • GPU power consumption analysis
  • Power Delivery Network (PDN) understanding
  • GPU architecture knowledge
  • Python for scripting and automation

Nice-to-have

  • Post-silicon power characterization experience
  • Experience with DVFS and AVFS
  • Understanding of PDN impedance targets

Key Requirements

  • MS or PhD in CS, EE, or CE (or equivalent experience)
  • 5+ years of experience in GPU kernel development
  • Strong CUDA and C++ programming skills
  • Experience with GPU performance profiling tools
  • Working knowledge of Power Delivery Networks (PDNs)

Work Rights

Not specified

Tailored Resume

Cover Letter