Senior Datacenter GPU Power Architect

Job Title

Senior Datacenter GPU Power Architect

Role Summary

Design and validate power and performance estimation methodologies and tools for datacenter GPUs and systems. Lead early architecture exploration focused on Perf/Watt, set power targets, and influence product roadmap decisions across GPU, platform, and datacenter levels.

Work with architects, designers, VLSI, software, and system teams to enable energy-efficient GPU and datacenter solutions.

Experience Level

Senior-level. Targeted experience: 8+ years in power/performance modeling, estimation, or related roles.

Responsibilities

Primary responsibilities include building models, defining KPIs, and driving analysis to improve energy efficiency at chip and datacenter scale.

• Develop power estimation models, flows, and tools for GPUs and datacenter systems (e.g., DGX/HGX).
• Perform early architecture and system-level exploration with emphasis on energy efficiency and TCO.
• Set and track power targets across ASIC milestones; perform performance-vs-power tradeoff analysis to inform roadmap decisions.
• Apply machine learning techniques to create accurate power and performance models for GPUs, CPUs, switches, and platforms.
• Analyze GenAI/HPC workload characteristics at datacenter scale to identify HW/SW features that improve Perf/Watt.
• Perform pre-silicon thermal analysis, including thermal use-case design and power mapping for thermal simulation.
• Model and analyze advanced technologies such as chiplets and high-speed/high-density interconnects.
• Collaborate cross-functionally to communicate findings and influence architecture and product decisions.

Requirements

Must-have technical skills and competencies. Preferred items are noted.

• Strong fundamentals in energy-efficient chip and system design and tradeoffs.
• Familiarity with low-power techniques: multi-VT, clock gating, power gating, DVFS.
• Deep understanding of processor (GPU preferred), server, and system-software architectures and their performance/power modeling approaches.
• Experience with Python and silicon data analysis workflows.
• Familiarity with performance monitors, simulators, and related architecture analysis tools.
• Ability to deploy ML techniques for modeling and analysis of power/performance data.
• Strong communication, collaboration, and organizational skills for cross-functional work.

Education Requirements

MSEE or MSCE preferred; PhD is preferred for some candidates. The posting also accepts equivalent practical experience in power/performance estimation and optimization. Relevant fields include Electrical Engineering, Computer Engineering, Computer Science, or related technical disciplines.

About the Company

Company: NVIDIA

Headquarters: Santa Clara, California, USA

NVIDIA is a global leader in accelerated computing, renowned for its innovative solutions in AI and digital twins that transform diverse industries. The company specializes in networking technologies, providing end-to-end InfiniBand and Ethernet solutions for servers and storage that optimize performance and scalability. NVIDIA serves sectors such as high-performance computing, enterprise data centers, and cloud computing, constantly reinventing its products and services to stay ahead in the market.