Aditi Krishnapriyan

Timeline

1. 2010–2013

   Chemistry and physics at UCSB

   A native of Des Moines, Iowa, she earns a bachelor's degree in chemistry and physics at UC Santa Barbara in 2013 after adding physics to her planned chemistry studies. She joins a condensed-matter and solid-state physics lab, where she uses fundamental equations to model materials.

2. By 2020

   A Stanford PhD and research

   By 2020, she completes a PhD in computational condensed matter physics and materials science at Stanford, as a member of its Materials Computation and Theory Group. During graduate school, she works on machine learning research at Google and Toyota Research Institute; in 2016, she completes a Department of Energy Computational Science Graduate Fellowship practicum at Los Alamos National Laboratory, optimizing density-functional tight-binding parameterizations with machine learning.

3. 2020–2022

   Alvarez fellowship at Berkeley Lab

   She joins Lawrence Berkeley National Laboratory's Computational Research Division, in its Data Analytics and Visualization Group, as a Luis W. Alvarez Fellow in Computing Sciences in 2020 and holds the fellowship through 2022. One of her first projects uses topology and statistical mechanics to study condensed matter systems.

4. 2021–2022

   Physics-informed and continuous models

   In 2021, she coauthors a study showing that physics-informed neural networks can fail to learn relevant physical phenomena on slightly more complex problems. In 2022, she coauthors a study showing that models tested for continuous dynamics can deliver better interpolation and extrapolation.

5. January 2024

   Faculty at UC Berkeley

   She joins UC Berkeley's faculty as an assistant professor of Chemical and Biomolecular Engineering in January 2024, developing machine learning methods that solve partial differential equations in scientific simulations.

6. 2025

   A DOE Early Career Award

   She receives the Department of Energy Early Career Award in 2025 for machine learning methods that scale with dataset size to model complex, multi-scale phenomena. The resulting tools can support studies of new battery electrolytes and molecular interactions.

7. August 2026

   An NSF CAREER Award

   She receives the NSF CAREER Award in August 2026 for physics-inspired machine learning frameworks that learn generalizable representations from large datasets and use empirical neural scaling laws to guide model design.

8. As of October 2026

   Berkeley faculty and lab scientist

   As of October 2026, she is an assistant professor in UC Berkeley's Chemical and Biomolecular Engineering and Electrical Engineering and Computer Sciences departments and a faculty scientist in Berkeley Lab's Applied Mathematics and Computational Research Division. Her research group develops machine learning methods for physical systems and scientific simulations, including generative models, surrogate models for expensive simulations, and reliable prediction under distribution shift.