Research Scientist, Efficient Deep Learning

About the role

NVIDIA is looking for a researcher to join the deep learning efficiency group to advance model optimization techniques. You will focus on creating methods that improve performance while maintaining practical utility for real-world applications.

Key facts

What you'll do

• Develop and execute new strategies for efficient deep learning.
• Produce and publish original research findings.
• Work alongside internal team members and external researchers.
• Provide mentorship to interns.
• Represent the company by presenting at industry events and conferences.
• Partner with product divisions to integrate new technologies into actual offerings.

Requirements

• Ph.D. in Computer Science, Electrical Engineering, or a related field, or equivalent research experience.
• Proficiency in the theoretical and practical aspects of deep learning and computer vision.
• Proven experience with large language models and large vision-language models.
• Strong programming ability in Python and PyTorch.
• Practical experience in large-scale model training, including data preparation and model parallelization techniques like tensor and pipeline parallelism.
• Demonstrated history of high-quality research output.
• Strong communication skills.

Nice to have

• Background in pruning, quantization, neural architecture search, or efficient backbone design.
• Proficiency in C++ and parallel programming frameworks such as CUDA.

Skills & tools

• Python
• PyTorch
• C++
• CUDA
• Large-scale model training
• Model parallelization

Practical notes

• Base salary range: 168,000 USD to 264,500 USD, depending on experience, location, and internal pay parity.
• Compensation includes equity and benefits.
• Application deadline: June 15, 2026.
• NVIDIA utilizes AI tools during the recruitment process.
• The company is an equal opportunity employer and does not discriminate based on protected characteristics including race, religion, gender, age, disability, or veteran status.