Distinguished Engineer, Power Architecture

What you'll do

• Set cross-portfolio, multi-generation power strategy and guide performance-versus-power trade-offs and product roadmaps.
• Lead silicon and vendor/IP co-architecture and end-to-end thermal and system-level power management.
• Drive ML- and workload-aware modeling, post-silicon optimization, technical leadership, and industry influence.

What they ask for

• MSEE, MSCE, or PhD, or equivalent experience demonstrating significant impact in a relevant field.
• Demonstrated ownership of power architecture across several successful silicon generations.
• Deep applied expertise in low-power and energy-efficient design, and authoritative knowledge of modern compute architectures, interconnects, memory, and system layers.
• Proven development and implementation of power/performance modeling methods, metrics, and tools, plus experience architecting power solutions with internal teams and external partners.

Nice to have

• Power and thermal architecture for a high-volume GPU, accelerator, SoC, or server SoC; or model-to-hardware co-optimization for ML, perception, or graphics workloads.
• Co-design experience with platform power delivery and cooling; applied ML research or production work in power/performance; or deep implementation expertise and technical mentorship.

Pay

320,000 USD - 488,750 USD (base salary)

Tools

GPU, CPU, NVLink, PCIe, Ethernet, HBM, LPDDR, on-die SRAM

AI focus

Drives ML- and workload-aware modeling and post-silicon optimization; preferred experience includes ML workload co-optimization and applying ML to power/performance modeling, workload characterization, or DVFS/AVFS strategies. NVIDIA also uses AI tools in recruiting.