Taro Logo

Deep Learning Performance Architect

NVIDIA is the world leader in accelerated computing, pioneering solutions in AI and digital twins that transform industries.
Machine Learning
Staff Software Engineer
In-Person
5,000+ Employees
5+ years of experience
AI

Job Description

NVIDIA, the world leader in accelerated computing, is seeking a Deep Learning Performance Architect to join their AI performance modeling team. This role focuses on developing and optimizing deep learning systems, particularly for LLM workloads, on state-of-the-art hardware architectures. The position offers an opportunity to work at the intersection of hardware and software optimization for AI systems, making significant contributions to NVIDIA's next-generation inference products.

The ideal candidate will analyze cutting-edge deep learning networks, develop analytical models, and work on performance optimization across both hardware and software domains. They will collaborate with architecture, software, and product teams to influence the direction of future deep learning solutions. This role requires extensive experience with AI models, particularly LLMs and AIGC models, along with deep knowledge of machine learning frameworks and hardware architectures.

NVIDIA offers a competitive compensation package and is known for being one of the technology industry's most desirable employers. The company maintains a strong commitment to diversity and inclusion, fostering an innovative and collaborative work environment. This position provides an exceptional opportunity to work on groundbreaking technology that is transforming industries while being part of a forward-thinking team at a company that continues to push the boundaries of what's possible in AI and accelerated computing.

Last updated 13 days ago

Responsibilities For Deep Learning Performance Architect

  • Analyze state of the art DL networks (LLM etc.), identify and prototype performance opportunities
  • Develop analytical models for deep learning networks and algorithms
  • Specify hardware/software configurations and metrics to analyze performance, power, and accuracy
  • Collaborate across teams to guide next-gen deep learning HW/SW direction

Requirements For Deep Learning Performance Architect

Python
  • BS, MS or PhD in relevant discipline (CS, EE, Math, etc.) or equivalent experience
  • 5+ years work experience
  • Experience with popular AI models (e.g., LLM and AIGC models)
  • Familiar with deep learning frameworks (Torch/JAX/TensorFlow/TensorRT)
  • Knowledge and experience on hardware architectures for deep learning applications

Benefits For Deep Learning Performance Architect

  • Competitive salaries
  • Generous benefits package