Energy Modeling Engineer – ML Power for GPUs

NVIDIA Gruppe; Santa Clara (CA); On-site; USD 116,000 - 190,000; Full time

Job summary

NVIDIA is seeking an Architecture Energy Modeling Engineer in Santa Clara, CA to help design energy-efficient GPUs. You will build ML-based power models, work with architects, ASIC/design teams, and performance engineers to study energy consumption and drive energy-aware optimization across the design cycle.

The role emphasizes building energy models integrated into simulators and silicon platforms, with opportunities to impact architectural and power-management decisions.

Qualifications

• Pursuing or completed MS/PhD in EE/CE/CS or equivalent experience.
• Strong coding skills in Python and C++.
• Background in ML/AI and statistical modeling.
• Background in computer architecture and energy-efficient GPU design.
• Familiarity with Verilog and ASIC design principles is a plus.
• Ability to analyze algorithms and discuss runtime/memory complexities.
• Understanding of energy consumption concepts and low-power design.
• Desire to apply quantitative decision-making to improve energy efficiency.
• Good verbal and written communication and interpersonal skills.

Responsibilities

• Develop an energy-efficient GPU through collaboration with architects, designers, and performance engineers.
• Identify design features and workloads for ML-based unit power/energy models.
• Develop workflows to train models using ML and statistical techniques.
• Improve model accuracy with different representations, objectives, and learning algorithms.
• Estimate data movement power/energy accurately across design stages.
• Correlate model energy with silicon-level estimates to bridge early estimates.
• Integrate power/energy models into performance platforms for combined reporting.
• Prototype new architectural features and build energy models to analyze system impact.
• Suggest studies and solutions to improve GPU perf/power.

Skills

Python

C++

Machine Learning

Statistical modeling

Computer architecture

Algorithms analysis

Energy estimation basics

Quantitative decision-making

Communication skills

Education

MS or PhD in Electrical Engineering, Computer Engineering, Computer Science

Tools

Verilog

Job description

NVIDIA is seeking an Architecture Energy Modeling Engineer in Santa Clara, CA to help design energy-efficient GPUs. You will build ML-based power models, work with architects, ASIC/design teams, and performance engineers to study energy consumption and drive energy-aware optimization across the design cycle.

The role emphasizes building energy models integrated into simulators and silicon platforms, with opportunities to impact architectural and power-management decisions.