In my free time, I enjoy reading astronomy books, painting, and working as a voice actor for children's audiobooks on YouTube!

Minseo (Sonia) Kim

I am an incoming Ph.D. student in Electrical Engineering at Stanford University, starting in Fall 2026. I am currently a second-year M.S. student in the Stanford Computational Imaging Lab, advised by Prof. Gordon Wetzstein.

I am broadly interested in developing computational imaging algorithms that tackle complex inverse problems across diverse scientific domains such as medical imaging, optical imaging, astronomical imaging, and structural biology, leveraging AI tools like generative models and physics-informed neural networks to enhance reconstruction. I am passionate about bridging the gap between computational methods and real-world imaging challenges to advance scientific discovery.

I did my B.S.E. (Honors) in Electrical Engineering and Data Science with a minor in Mathematics at University of Michigan - Ann Arbor, where I was advised by Prof. Jeffrey Fessler.

Research Projects

A diffusion model-based solver for inverse problems that is better, faster, and more robust to noise than relevant baselines.

I implemented a score-based diffusion model with posterior sampling to reconstruct high-quality MRI images from undersampled k-space data, tested on over 3000 FastMRI dataset images.

Designed the double interstellar dust scattering physics model and developed mathematical proofs to derive halo intensity using analytic and numerical methods.

Implemented 2D branchless distance-driven forward projection and backprojection algorithm for computed tomography (CT) reconstruction using both PyTorch and Julia language.

Industry Experiences

Research Intern
SLAC National Accelerator Laboratory, MLCV@LCLS
June 2025 -

Research in the area of computational structural biology focused on developing experiment-grounded protein structure generative models, with Doris Mai, Frederic Poitevin, and Prof. Matthias Kling.

Digital Signal Processing Intern
Zoox, Inc., Advanced Hardware Engineering Team
May 2024 - August 2024

Worked on developing AI-powered audio mining systems and multimodal data processing pipelines, implementing models like PANN and AudioCLIP for text-based audio search and applying DSP techniques for vehicle data preprocessing, with Venkata Chebiyyam and Shaminda Subasingha.

Others