Senior Research Scientist, Efficient Deep Learning

NVIDIA · Semiconductors · Santa Clara, CA

Senior Research Scientist at NVIDIA focusing on efficient deep learning methods, including post-training optimization, architecture design, and resource-efficient training/fine-tuning. The role involves research, implementation, publication, collaboration, and technology transfer to products.

What you'd actually do

  1. Research, design and implement novel methods for efficient deep learning.
  2. Publish original research. Speak at conferences and events.
  3. Collaborate on research with internal team members, internal teams as well as external researchers. Mentor interns.
  4. Work with product groups to transfer technology.

Skills

Required

  • Ph.D. in Computer Science/Engineering, Electrical Engineering, or equivalent experience
  • 3+ years relevant post-graduate research experience
  • Excellent knowledge of theory and practice of computer vision methods, as well as deep learning
  • pruning
  • quantization
  • NAS
  • efficient backbones
  • large-scale model training
  • data preparation
  • model parallelization (tensor and pipeline)
  • Python
  • PyTorch
  • C++
  • parallel programming (e.g., CUDA)

Nice to have

  • Experience with large language models
  • Experience with large vision-language models

What the JD emphasized

  • background in pruning, quantization, NAS, efficient backbones is required
  • Hands-on experience with large-scale model training including data preparation and model parallelization (tensor and pipeline) is required
  • An outstanding research track record

Other signals

  • novel methods for efficient deep learning
  • post-training model optimization
  • efficient architecture design
  • adaptive/dynamic inference
  • resource-efficient training and fine-tuning