Luke Pfister

I'm currently a Machine Learning Engineer on the Modeling Foundation & Co-Design team at Meta. I work on making large-scale recommendation systems more efficient.

Previously, I was a staff scientist at Los Alamos National Laboratory. I was focused on in the intersection of computational imaging, the theory of inverse problems, and machine learning. I spent most of my time as the lead developer for SCICO, a Python library for computational imaging and inverse problems powered by JAX.

I had a brief stint as a postdoc in the Chemical Imaging and Structures Group at the University of Illinois at Urbana-Champaign.

I completed my PhD under Yoram Bresler, also at the University of Illinois at Urbana-Champaign. I worked closely with Rohit Bhargava and P. Scott Carney. My dissertation was split between the design of fast sparsifying transforms and efficient methods for spectroscopic tomography.

I received the Andrew T. Yang Research Award in 2014 and 2015 to fund the application of modern signal processing and model based image reconstruction to spectroscopic optical tomography.

During the summer of 2015 I was an intern with the Computational Sensing & Imaging group at Rambus Labs. I worked on the design of and algorithms for application-specific lensless imaging systems.