Research grounded in curiosity, statistics, and engineering.

Finding structure in how we communicate

My path into speech research began with computer science, expanded through graduate study in statistics, and developed through applied work across large-scale software, machine learning, and conversational AI.

At NVIDIA, my current work focuses on applied AI agents that can reason through data science tasks and work with analytical tools. Previously, I researched text-to-speech, speech-to-speech language models, neural audio codecs, and parameter-efficient adaptation.

Before NVIDIA, I worked at Google and Microsoft and conducted research at the University of Illinois Urbana-Champaign on OCR correction, language modeling, and computational stylometry.

Outside research, I enjoy sketching, playing football, and following ideas that make difficult technical concepts feel intuitive.

Views expressed here are my own and do not represent the organizations with which I am affiliated.