Levi Harris

Hey there! My name is Levi. I am a first-year master’s student and researcher at the University of North Carolina at Chapel Hill working under the supervision of Dr. Tianlong Chen at the UNITES lab. I am broadly interested in problems related to AI4Science. To explain why, I’ve attached one of my favorite answers to the question “What is Artificial Intelligence?” below.

“Modern AI is a horizontal enabling layer. It can be used to improve everything. It will be in everything. This is most like electricity.”

• Jeff Bezos

This quote gets to the heart of why the field of AI has grown to enormous popularity in recent years. AI, at its core, is a universal process improving substrate. AI is horizontal because it slots into almost any technology stack. My primary aim as a researcher is to find ways to fit AI into important, well-established processes by developing methods that are one, nonintrusive, and two undeniably better than the status quo.

My current work seeks to develop new image-processing techniques for atomic force microscopy (AFM), particularly for fast characterization of two-dimensional (2D) materials like MoS2. My colleagues and I hope that by bridging the gaps between state-of-the-art computer vision methods and well-established Metrology techniques, we can breathe new life into old hardware.

Previously, during my undergrad at UNC, I worked with Dr. Gedas Bertasius to build pipelines for structured video generation and human behavioral modeling. Together we also built massive, (i.e., 30TB+) densely annotated datasets to push video understanding models to new limits.