Massive stars represent the extremes of stellar evolution, and are the sites of key physical processes that shape the Universe. However, their short lifetimes and intrinsic rarity make them a unique challenge to study. As a result, a number of open questions still exist: What sequence of evolutionary states to massive stars evolve through? What are the roles of rotation and binary interactions in the evolution of massive stars? Which stars end their lives in explosive supernovae, and which collapse directly into black holes? What do massive stars look like in the moments before their deaths?

Answering these questions requires high quality observations of large numbers of massive stars. It is only recently that all-sky missions like TESS and Gaia have provided us with such observations. At last, we can take techniques that are well-tested on low mass stars, bring them to bear on these fundamental problems in stellar evolution, and make statistically robust conclusions.  

In my work, I use big data techniques to identify large samples of massive stars, which I can then study in detail. For example, I use TESS observations to characterize microvariability in evolved massive stars. Using a cutting-edge method incorporating Gaussian process regression, I identified periodically variable stars, and discovered a new class of pulsating star, Fast Yellow Pulsating Supergiants, or FYPS. The properties of FYPS are consistent with "post red supergiant" objects, and can help us solve the long-standing red supergiant problem. I am also interested in the application of machine learning to astronomical data, and recently developed a classification technique that can identify rare classes of massive stars that would otherwise require dedicated telescope time to find. This technique will become essential in the era of JWST, the Nancy Grace Roman telescope, and the Legacy Survey of Space and Time.