GPU Power Architect for AI & Energy Efficiency

NVIDIA Corporation; Santa Clara (CA); On-site; USD 124,000 - 242,000; Full time

Benefits offered by this job

Equity

Benefits package

Job summary

NVIDIA Corporation is seeking a researcher/engineer to architect power features and system-level power management for its GPU roadmap. You will work with architects, software, and ASIC teams to study and implement strategies addressing energy efficiency across DL training, ADAS, gaming, video playback, and idle scenarios.

The role involves deploying ML-based power and performance models for GPUs and platforms, including using Python and ML frameworks, and leveraging lab measurements for

Qualifications

• MS or PhD in Electrical or Computer Engineering (or equivalent experience).
• Knowledge of performance simulators/monitors and low power architectures/techniques.
• Experience with lab setup and measurement using equipment like scope/DAQ.
• Exposure to tools/flows such as Design Compiler, PTPX, and Power Artist (stand-out).

Responsibilities

• Architect GPU power features and system-level power management for NVIDIA products.
• Collaborate with Architects, Software Engineers, ASIC Design Engineers, and Product teams to define power strategy for GPU roadmap.
• Research and develop energy-efficiency models for DL training, ADAS, gaming, video playback, and idle use-cases.
• Deploy ML techniques to build accurate power/performance models for GPUs and platforms.

Skills

Low Power Architecture

Power Modeling

Performance Simulators

Lab Measurement

Education

MS or PhD in Electrical/Computer Engineering

Tools

Python

TensorFlow

Pandas

NumPy

PyTorch

Job description

NVIDIA Corporation is seeking a researcher/engineer to architect power features and system-level power management for its GPU roadmap. You will work with architects, software, and ASIC teams to study and implement strategies addressing energy efficiency across DL training, ADAS, gaming, video playback, and idle scenarios.

The role involves deploying ML-based power and performance models for GPUs and platforms, including using Python and ML frameworks, and leveraging lab measurements for