Zack Berger

Welcome!

I'm a third-year PhD student at MIT EECS, advised by John Guttag and Collin Stultz and supported by the DoD NDSEG Fellowship.

My work aims to make proactive health management accessible to everyone by developing AI systems that monitor health, anticipate risk, and can be deployed safely and reliably.

Previously, I was a software engineer at Google working on generative AI infrastructure. I completed my BS in computer science at UCLA, where I worked with Amit Sahai and was a research assistant with the Vision Lab.

I'm also a jazz pianist and comedian! Check me out on Moonshine and Vibe Box.

Contact: zberger [at] mit [dot] edu

Current Projects

TinyAF. Developing and deploying a compact model for continuous, on-device detection of atrial fibrillation and atrial flutter on wearables.
PACE-VD. Developing a model that uses paced ECGs to forecast systemic ventricular dysfunction in children with congenital heart disease.

Papers

Position: Evaluation of ECG Representations Must Be Fixed
Zachary Berger, Daniel Prakah-Asante, John Guttag, and Collin M. Stultz.
ICML, 2026 [Journal] [Website] [arXiv] [Code]

Estimating CardioMEMS Pulmonary Artery Diastolic Pressure With a Non-invasive Cardiac Hemodynamic Artificial Intelligence monitoring System (CHAIS)
Zachary Berger, Roey Ringel, Samuel Roytburd, Nir Ayalon, Deepa Gopal, Lana Tsao, John Guttag, and Collin M. Stultz.
AHA Scientific Sessions, 2025 [Abstract] [Talk]

DOCCI: Descriptions of Connected and Contrasting Images
Yasumasa Onoe, Sunayana Rane, Zachary Berger, Yonatan Bitton, Jaemin Cho, Roopal Garg, Alexander Ku, Zarana Parekh, Jordi Pont-Tuset, Garrett Tanzer, Su Wang, and Jason Baldridge.
ECCV, 2024 [Website] [arXiv]

Stereoscopic universal perturbations across different architectures and datasets
Zachary Berger, Parth Agrawal, Tian Yu Liu, Stefano Soatto, and Alex Wong.
CVPR, 2022 [Journal] [arXiv] [Code]

Industry Experience

Software Engineer
Participated in image generation research and worked on serving generative AI at scale, contributing to projects including Imagen, MusicLM, and logo recognition.

ML Engineer Intern
Developed and deployed a streaming system for detecting corrupted vehicle data across a multi-petabyte dataset.

Teaching

I taught a seminar called Demystifying Computer Science, an approachable end-to-end look at CS principles for non-STEM students. The lectures are available on YouTube.