Senior Research Scientist, Nemotron Post-training

What you'll do

• Develop and implement post-training technologies for Nemotron models.
• Advance open-source foundation models by creating training data and benchmarks.
• Solve large-scale challenges in model lifecycle management and deployment.
• Engage in synthetic data research and algorithmic development for agentic RL.
• Collaborate on vendor data acquisition and experimentation for model improvement.

What we're looking for

• Master or PhD in computer science, machine learning, or quantitative fields.
• 5+ years experience in model post-training, reinforcement learning, and agentic systems.
• Hands-on data curation and model training for Agentic and Reasoning capabilities.
• In-depth experience with inference and deployment environments like vLLM, SGLang, TRT-LLM.
• Industrial experience in reinforcement learning for leading foundation models.
• Experience optimizing model quality using real-world traffic feedback.