Kaylee de Soto

Hi! I’m currently a PhD candidate at the Center for Astrophysics at Harvard University, working under the direction of Dr. Ashley Villar. I am a LSST-DA Data Science Fellow, a former LINCC Frameworks Incubator PI, and an honorable mentionee from both the Ford Foundation Predoctoral Fellowship and the Chambliss Student Award. Previously, I obtained my master’s degree in astrophysics from Penn State in 2023, and my bachelor’s degree in physics and math with computer science from MIT in 2021.

My research focuses on characterizing populations of supernovae from their photometry. I have developed pipelines for real-time photometric classification of supernovae that use both parametric fits (e.g. Superphot+) and low-dimensional auto-encodings (e.g. SupernoVAE). Currently, I am studying our ability to constrain Type IIP supernova progenitor distributions from ZTF photometry, and improving supernova population isolation within the DECam deep drilling fields. You can read more about my current projects here.

When the Vera C. Rubin observatory begins its Legacy Survey of Space and Time in late 2025, it is expected to detect about a million supernovae per year! I am passionate about optimizing our current follow-up infrastructure for this rapid datastream, by working closely with data brokers like ANTARES and by incorporating state-of-the-art machine learning techniques. I additionally work on architectures that can combine supernova information from multiple surveys and modalities, to best integrate Rubin into the existing observational ecosystem.

If interested in collaboration (or in contributing to my active codebases!), I am most easily reachable via email (kaylee.de_soto@cfa.harvard.edu).