The Data Science Institute is excited to welcome the fourth cohort of Eric and Wendy Schmidt AI in Science Fellows. A program of Schmidt Sciences, the Fellowship provides transformative training and support to exceptional early-career researchers to accelerate discovery across the natural sciences through innovative applications of machine learning and artificial intelligence.

Fellows span departments from molecular biology to particle physics to climate sciences, each bringing their own unique expertise while sharing a common vision: harnessing AI to push the boundaries of scientific understanding in their respective fields. 

“The diversity of expertise demonstrates AI’s potential to transform scientific inquiry across domains,” said Rebecca Willett, Faculty Director of AI at the Data Science Institute and Worah Family Professor in the Wallman Society of Fellows in the Department of Statistics and Computer Science and the College. “These Fellows are positioned to make groundbreaking contributions to the way we conduct research in the natural sciences.”

The interdisciplinary nature of the program creates opportunities for cross-pollination of ideas, methods, and collaboration. Fellows benefit from the University of Chicago’s data science ecosystem and excellence in research, including access to cutting-edge computational resources and mentorship from world-class faculty across departments.

From understanding marine ecology to discovering new physics, these scholars represent the next generation of researchers advancing scientific discovery through the thoughtful integration of AI and domain expertise.

Meet the Fourth Cohort of Schmidt AI in Science Fellows

Coralie Rousseau joins from the Station Biologique De Roscoff, where she earned her Ph.D. studying the ecological roles of microbial communities. Rousseau will be working with Professors Cathy Pfister (Ecology and Evolution) and Claire Donnat (Statistics) as she investigates how the thousands of microbial species within marine “holobionts” (the dynamic functional groups formed by hosts and microbes) interact to maintain host health. Her research aims to reduce the dimensionality of complex microbial data, map intricate ecological interactions, and forecast community shifts that could impact the resilience of marine ecosystems.

Didem Cifci brings her expertise in computational pathology to address head and neck squamous cell carcinoma. Having completed her Ph.D. at RWTH Aachen University under Jakob N. Kather, Cifci will work with Assistant Professor Alex Pearson (Medicine) to develop machine learning approaches that can classify molecular states from diagnostic slides. She uses multimodal AI methods to characterize the molecular and histological markers of therapeutic response to advance precision treatment strategies.

Emma Liu’s climate science work combines radar observations, ice core data, and physics-based simulations to understand ice rheology across vastly different scales.  Emma earned her Ph.D. at Stanford under Jenny Suckale. Liu will be joining Assistant Professor Meghana Ranganathan in Geophysical Sciences to advance our ability to model and predict glacier flow in a changing world.

Konstantin Gerbig’s research explores planet formation, applying machine learning to understand dust diffusion in protoplanetary disks—a process that determines how planets begin to form. His work aims to shed light on the earliest stages of planetary system evolution. Gerbig recently completed his Ph.D. at Yale with Greg Laughlin, and at UChicago will join Assistant Professor Diana Powell‘s group in Astronomy and Astrophysics.

Marco Biroli’s work is focused on non-equilibrium statistical physics. At UChicago, he will be joining Professor Vincenzo Vitelli (Physics) as he works to develop physically inspired machine learning models to devise algorithms that respect underlying physical constraints (e.g., symmetries, conserved quantities, stochastic dynamics), thus improving interpretability, generalization, and robustness. He completed his Ph.D. at the Université Paris-Saclay, Laboratoire de Physique Théorique et de Modèles Statistiques.

Radha Mastandrea is researching and defining why physics-aware machine learning architectures perform so well in particle physics. Working with Assistant Professor Karri DiPetrillo (Physics), she hopes to extrapolate these advances to broader neural network classification tasks. Prior to coming to UChicago, she completed her Ph.D. at the University of California, Berkeley with Ben Nachman.

Seongsoo Kim investigates the cellular architecture and mechanics of the biopolymer actin. Kim, who will continue his research in the lab of Professor Margaret Gardel (Physics), combines experimental techniques with machine learning to understand how phase-separated droplets influence the assembly and mechanical properties of actin networks. Prior to coming to UChicago, Kim completed his PhD in materials physics at Harvard University.

Siddarth Achar will continue to tackle environmental contamination through molecular design with Associate Professor Andrew Ferguson (Molecular Engineering) and Professor Junhong Chen (Molecular Engineering). Achar develops statistical methods that leverage AI to design molecular probes for detecting PFAS “forever chemicals” in drinking water. His Ph.D. in computational modeling at the University of Pittsburgh provided the foundation for this application for environmental health.

Robert Serafin applies computer vision and machine learning to decode the spatial organization of tumor immune microenvironments. Working in the Department of Hematology and Oncology under the mentorship of Assistant Professor (and former Schmidt Fellow) Madeleine Torcasso and Associate Professor Akash Patnaik Serafin will analyze microscopy data to understand how interactions between tumor and immune cells influence treatment response. He aims to derive spatial biomarkers to develop clinical decision-making tools for precision oncology. Prior to coming to UChicago, Serafin worked to develop 3D microscopy pipelines at the Allen Institute for Brain Science and completed his Ph.D. in computational pathology at the University of Washington. 

These nine scholars exemplify the program’s commitment to advancing research across domains. The fourth cohort joins a growing community of AI-trained scientists reshaping how we approach fundamental questions across the natural sciences.