Senior Research Scientist at NVIDIA's Applied Deep Learning Research Group
Adjunct Professor at Boston University
Research focuses on advancing large language models (LLMs) by enhancing reasoning capabilities and ensuring safety through mitigation of toxicity and bias
Lead contributor to the Nemotron family of models, with extensive work on data curation, pretraining, and scaling
Currently optimizing pretraining pipelines via data selection, blending, and ordering strategies to maximize downstream accuracy
Particularly interested in improving LLM reasoning, including synthetic data generation for advanced mathematical reasoning and enabling models to handle longer, complex reasoning tasks