Summary
Victor Shepardson is a PhD researcher and interdisciplinary machine learning engineer with 12 years of experience building generative systems for audio, speech, and video. Based in Ann Arbor, he combines an MA in Digital Musics and a CS BA to design AI-driven musical instruments and custom synthesis tools, bridging creative practice and deep learning research. His work spans industry and academia—from deep generative sequence models and structural causal approaches at Ntrepid to self‑employed projects delivering bespoke speech/video synthesis for artists. Currently conducting doctoral research in Icelandic institutions, he brings hands‑on PyTorch expertise, product-minded engineering, and a history of teaching complex technical subjects to creatives and students. An unusual thread through his career is the consistent focus on sonically informed ML applications, making him fluent in both artistic workflows and rigorous model development.
12 years of coding experience
4 years of employment as a software developer
Master of Arts (M.A.), Digital Musics, Master of Arts (M.A.), Digital Musics at Dartmouth College
Bachelor of Arts (B.A.), Computer Science, Bachelor of Arts (B.A.), Computer Science at University of Virginia
English