Sean Hardesty Lewis

I am a predoctoral researcher at MIT working on embodied AI, world models, and scene understanding for large-scale physical and virtual environments.

Previously, I completed my M.S. in Information Science at Cornell Tech, where I was advised by Wendy Ju and Anthony Townsend. There, I created Searchable.City, a semantic world model built from over 50 million street-view images that enables natural-language search of the world.

I also created Human Operator, a vision-language decision support system for embodied AI that explored AI-guided human embodiment, reaching over 30 million views online.

Before Cornell, I received my B.S. in Mathematics from UT Austin, where I worked with Junfeng Jiao on AI systems for city-scale sensing and simulation.

I am interested in building AI systems that allow agents to understand, simulate, and act within complex 3D environments.