Info

I currently work at Apple as a Senior Machine Learning Engineer on the Maps 3D Vision team. We make the visually stunning maps for the Detailed City Experience product available in over 40 cities around the world. We develop algorithms to go from pixels to beautiful and centimeter accurte 3D map data at city, and even country scale for roads, buildings, trees, lane markings, crosswalks, and much more.

Before that I spent four years at Third Wave Automation on autonomous forklifts. As head of Perception and Manipulation, I led the development of a real-time vision stack for object detection, pose estimation, planning, and manipulation. If you thought autonomous driving was hard, try having the vehicle pick up something half its size weighing 3000 pounds, lift it 30 feet into the air, then place it with centimeter tolerances.

At the University of Washington, I was part of Dieter Fox's RSE Lab and Ali Farhadi's RAIVN Lab. I researched teaching machines to see and act in interactive visual environments. I also had research internships at Facebook AI Research with Dhruv Batra on Habitat and The Allen Institute for AI with Roozbeh Mottaghi and Ani Kembhavi on Prior.

A thread running through all of it is interest in representation learning - what makes features that generalize. Whether that is video as a source of temporal unsupervised labeling, a factored perception and control model, or a recurrent tracking memory that updates itself online, I'm always thinking how data can be processed efficiently into meaningful higher-dimensional spaces.

In my spare time, I enjoy whittling little animal figurines, getting lost in nature, playing piano, and petting cats.