Deep Learning Scientist, LLM Training Datasets

@ NVIDIA

About the Role

NVIDIA is seeking a dedicated Deep Learning Scientist, LLM Training Datasets to innovate and engineer high-quality datasets for large language model training. This highly technical role requires deep expertise in machine learning, data science, and data engineering.

What You'll Be Doing

• Develop datasets for LLM pre-training, fine-tuning, and reinforcement learning.
• Design and implement data strategies including collection, cleaning, labeling and augmentation.
• Generate high-quality synthetic data for domain-specific and safety-critical use cases.
• Define annotation guidelines and curate labeled datasets for model alignment including RLHF.
• Conduct experiments to optimize LLM performance with SFT and RL techniques.
• Collaborate with ML researchers, data scientists, and infrastructure teams to integrate efficient ML workflows.

What We Need To See

• Master’s or PhD in Computer Science, Electrical Engineering or related field, or equivalent experience.
• 3+ years industry experience in dataset development and training generative AI models.
• Proficiency in Python programming and familiarity with ML frameworks such as PyTorch and TensorFlow Data.
• Experience with synthetic data generation, distributed training paradigms and performance evaluation of LLMs.
• Strong collaborative and communication skills along with excellent problem solving capabilities.

Ways To Stand Out

• Contributions to open-source data tools or research publications.
• Experience with cloud platforms and modern data storage systems.
• A passion for AI with continuous evaluation of new tools and methodologies.

Additional Information

With highly competitive salaries, equity, and a comprehensive benefits package, NVIDIA is renowned as one of the technology industry’s most desirable employers. Applications are accepted until September 30, 2025. NVIDIA champions diversity and equal opportunity in all its hiring practices.

Key skills/competency

Deep Learning, LLM, Datasets, Machine Learning, Data Engineering, Synthetic Data, Reinforcement Learning, PyTorch, TensorFlow, Distributed Training