Senior GPU Architect, Deep Learning

NVIDIA•Durham, NC

6d•$184,000 - $356,500

About The Position

NVIDIA's GPU Architecture group is seeking a Senior GPU and Deep Learning Architect to contribute to GPU architecture efforts. The group works to improve GPU architecture for deep learning training and inference, advance parallel programming models, and develop architectures for parallel processing. In this role, you will develop and enhance GPU architectural features that advance parallel programming models or parallel computing performance. You will work with architects and researchers to build simulators, map deep learning workloads to current and future hardware, and validate new architectural features. The position involves work spanning software simulators, RTL, and real silicon.

Requirements

MS in Computer Science, Electrical Engineering, or Computer Engineering, or equivalent experience.
Experience working with hardware targeted at deep learning or mapping deep learning algorithms to hardware.
At least 8 years of relevant industry experience in GPU or other parallel programming architectures, or equivalent experience.
Strong programming ability in C, C++, Perl, and Python.
Background in computer architecture, parallel processing, signal processing, and/or high-performance computing.

Nice To Haves

Knowledge of state-of-the-art deep learning algorithms and attention mechanisms.

Responsibilities

Design hardware features for future processing architectures targeting deep learning training and inference.
Advance the state of parallel computation.
Stay knowledgeable about future parallel programming models and their impact on hardware.
Develop software for hardware simulators, test infrastructures, metrics systems, and databases.
Collaborate on documenting, designing, developing analysis and simulation tools, and validating and verifying functional or performance models.
Develop tests, test plans, and testing infrastructure for new graphics or parallel processing architectures.