New Grad AI GPU Power Architect — Power-Efficient GPUs

NVIDIA; Austin (TX); On-site; USD 100,000 - 190,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA is seeking an AI GPU Power Architect for new college graduates to join the Applied Power Architecture team in Austin, TX. You will contribute to power estimation models for GPUs and DGX systems, focusing on energy efficiency and Perf/Watt for AI, HPC, and data center platforms.

We welcome candidates pursuing or who recently completed a Bachelors or Masters in EE/CE, with strong architecture knowledge and DVFS concepts.

Qualifications

• Pursuing or recently completed a Bachelor's or Master's in Electrical or Computer Engineering.
• Strong understanding of Computer Architecture and Digital Logic/VLSI Design.
• Proficiency with Python and data analysis packages (Pandas, NumPy, PyTorch).
• Knowledge of energy efficient chip design fundamentals and DVFS.

Responsibilities

• Contribute to power estimation models and tools for GPU products and systems like DGX.
• Explore early GPU/system architectures with focus on energy efficiency and TCO at GPU/datacenter level.
• Assist with Performance vs Power analysis for future NVIDIA products.
• Apply ML techniques to develop accurate power and performance models for GPUs, CPUs, switches, and platforms.
• Model workload characteristics for GenAI/HPC at datacenter scale to drive HW/SW features for Perf@Watt.

Skills

Computer Architecture

Digital Logic Design

VLSI Design

Python

Energy-efficient design

DVFS

Education

Bachelor's degree

Master's degree

Tools

Pandas

NumPy

PyTorch

Job description

NVIDIA is seeking an AI GPU Power Architect for new college graduates to join the Applied Power Architecture team in Austin, TX. You will contribute to power estimation models for GPUs and DGX systems, focusing on energy efficiency and Perf/Watt for AI, HPC, and data center platforms.

We welcome candidates pursuing or who recently completed a Bachelors or Masters in EE/CE, with strong architecture knowledge and DVFS concepts.