About me

Hi there! I am a postdoc at Stanford University, working with Gordon Wetzstein and Mert Pilanci in the Department of Electrical Engineering. My research involves machine learning, signal processing, and optimization for solving inverse problems in computer vision as well as medical and scientific imaging. I am currently supported by an NSF Mathematical Sciences Postdoctoral Research Fellowship. I completed my PhD in Electrical Engineering and Computer Sciences at UC Berkeley in May 2023, where I was advised by Ben Recht and supported by an NSF Graduate Research Fellowship.

I’m on the 2023-2024 academic job market.

Outside of research, I like to hike (the banner photo is one I took at Crater Lake National Park), garden, cook, paint, and read fantasy and historical fiction novels.

Contact info

The best way to reach me is by email, at sarafk at stanford dot edu. You can also find me on Google Scholar, LinkedIn, and GitHub.

Research interests

I am generally interested in foundations of machine learning, particularly connections to signal processing and optimization, and applications to inverse problems that arise in computer vision and computational, medical, and scientific imaging. Please have a look at my PhD dissertation and/or dissertation talk. Some of my current research interests are:

• Finding the “right” signal representation for solving inverse problems, balancing interpretability and expressivity with computational constraints
• Bridging theory and practice for inverse problems, e.g. nonlinear compressive sensing
• Finding the “right” priors for solving inverse problems, e.g. how to leverage data-driven priors without succumbing to distribution shift fragility
• Applications to medical and scientific imaging, including MRI, CT, and cryo-EM