Senior Applied Deep Learning Research Scientist, Efficiency

Details

Join the ADLR – Efficiency team to make deep learning faster and consume less energy. The team influences next-generation hardware to improve AI efficiency, works on the Nemotron series of models to develop highly efficient open-source deep learning models, and develops technology, software, and algorithms for optimizing neural networks during training and deployment.

Research topics include quantization, sparsity, optimizers, reinforcement learning, efficient architectures, and pre-training. The team is part of the Nemotron pre-training organization and collaborates across the company to improve NVIDIA GPUs as an efficient AI platform. The role focuses on understanding the root causes of efficiency challenges and developing new algorithms, numeric formats, and architecture improvements.

Responsibilities

• Research low-bit number representations and pruning, including their effects on neural network inference and training accuracy.
• Evaluate requirements of state-of-the-art neural networks and co-design future neural network architectures and optimizers.
• Innovate with algorithms that make deep learning more efficient while retaining accuracy.
• Open-source or publish algorithms for broad use.
• Run large-scale deep learning experiments and analyze the effects of efficiency improvements.
• Collaborate with teams working on hardware, software, and deep learning architectures.

Requirements

• PhD in artificial intelligence, computer science, computer engineering, mathematics, or a related field; equivalent experience in relevant areas may substitute for an advanced degree.
• At least 5 years of relevant industrial research experience.
• Familiarity with state-of-the-art neural network architectures, optimizers, and large language model training.
• Experience with modern deep learning training frameworks and/or inference engines.
• Fluency in Python and solid coding and software engineering practices.
• Proven publication track record and/or ability to run large-scale experiments.
• Strong interest in neural network efficiency.

Preferred Qualifications

• Experience with quantization, pruning, numerics, and efficient architectures.
• Background in computer architecture.
• Experience with GPU computing, kernels, CUDA programming, and/or performance analysis.

Compensation And Benefits

• Base salary range for Level 4: USD 192,000–304,750 per year.
• Base salary range for Level 5: USD 224,000–356,500 per year.
• Eligible for equity and benefits.
• Applications will be accepted at least until February 8, 2026.

NVIDIA is an equal opportunity employer committed to fostering a diverse work environment. NVIDIA uses AI tools in its recruiting processes.