About

I am an applied machine learning researcher and data scientist focused on complex human-centered signals: speech, audio, movies, gaze, behavior, and brain data. I like problems where the data are rich, messy, high-dimensional, and not easily reduced to a tidy benchmark.

I recently completed my PhD in the Harvard-MIT Speech and Hearing Bioscience and Technology (SHBT) Program, where I worked with Satra Ghosh in the Senseable Intelligence Group at MIT. My work sits at the intersection of applied ML, computational neuroscience, and evaluation: building datasets, extracting features, fitting models, and testing whether the results are reliable enough to support real conclusions.

At MIT and Harvard, I built large-scale predictive modeling workflows for naturalistic fMRI, multimodal feature extraction pipelines for movie stimuli, pediatric speech datasets for machine learning challenges, and research tools for annotating audiovisual data. A through-line across this work is turning ambiguous questions into measurable targets, reproducible pipelines, validation strategies, and interpretable evidence.

Before my PhD, I was a lab manager in the Robertson Lab at Dartmouth and worked with the Kanwisher Lab at MIT, where I designed VR eye-tracking and salience modeling pipelines to study visual attention. I also have a master’s degree in Digital Musics from Dartmouth, where I worked with Michael Casey on machine learning and music information retrieval approaches to stimulus reconstruction.

My broader research path has included Alzheimer’s disease clinical trials, deep-sea marine biology, microelectronics, and genomics. The domains have varied, but the core pattern has stayed consistent: I use quantitative methods to understand complex biological, behavioral, and physical systems.

Outside of work, I write and produce music, play tennis, and have taken an unusually high number of photos of a Tibetan spaniel named François.