Used Tools & Technologies
Not specified
Required Skills & Competences ?
Python @ 4 Algorithms @ 4 Parallel Programming @ 4 GPU @ 4Details
The NVIDIA GPU Architecture group is seeking a Senior GPU Power Architect to develop and enhance features in GPU architecture that advance power efficiency for graphics and deep learning workloads. You will work with architects and researchers to build simulators, create power models, define requirements, and design power-saving hardware features. This role involves work across architecture, software, and hardware, including simulators, RTL, and real silicon.
Responsibilities
- Work with chip architects to understand architecture concepts and high-level system requirements.
- Identify key areas for power improvements across architecture, software, and hardware and develop new features, defining requirements and dependencies across teams.
- Create power models of key GPU units to evaluate architectural tradeoffs in DL/ML (training/inference) and graphics workloads.
- Drive improvements to architecture development processes to improve execution capabilities.
- Design new hardware power-saving features for future graphics and parallel processing architectures.
- Stay knowledgeable about future parallel programming models and their impact on hardware.
- Learn and work on simulators, RTL, and real silicon.
Requirements
- MS in Computer Science, Electrical Engineering, Computer Engineering or equivalent experience.
- 7+ years of relevant industry experience in GPU or CPU architecture (or equivalent experience).
- Programming ability in C++ and Python.
- Background in hardware design, computer architecture, graphics algorithms, parallel processing, signal processing and/or high performance computing.
Benefits and Compensation
- Base salary ranges (dependent on level, location, and experience):
- Level 4: 184,000 USD - 287,500 USD
- Level 5: 224,000 USD - 356,500 USD
- Eligible for equity and benefits.
- NVIDIA is an equal opportunity employer and fosters a diverse work environment.
- Applications accepted at least until December 7, 2025.
Additional Information
- Hybrid working arrangement indicated (#LI-Hybrid).
- The role involves collaboration with other architects and researchers and may require cross-team coordination across architecture, software, and hardware domains.