I build AI that makes images and videos.

I lead media generation at Meta Superintelligence Labs, most recently the Muse Image and Muse Video models. Before that I led a team on omnimodels at Google DeepMind, and before that I co-led Sora at OpenAI. My PhD was at Berkeley AI Research, where I invented InstructPix2Pix, which lets you edit a photograph by telling it what to change in plain language. Earlier I worked on video generation at NVIDIA and on the camera in Google’s Pixel phones. In 2025, MIT Technology Review named me one of its 35 Innovators Under 35.

I photograph wildlife and landscapes. In high school some of my photos won awards from National Geographic, Nature’s Best Photography and the National Wildlife Federation. It’s a hobby now.

I also once performed at the Beacon Theatre in New York, where I won an international award for a cappella beatboxing.

My whole life I’ve made images, videos and music for the joy of it, and now I make the tools so anyone can.