Stars are born in dense with a broad range of masses. The most massive ones evolve quickly, ending their lives in as little as a few million years, leaving behind black holes as remnants. The globular clusters seen in all galaxies are thought to contain some of the oldest stars in the Universe. This research team will combinine astronomy, theoretical astrophysics, high-performance computing, and machine learning to develop state-of-the-art computer models of star clusters, simulating their evolution since the earliest stages of star formation in the distant Universe. The focus of this work will be on the dynamical interactions of black holes with other stars, through which they form compact binaries (e.g., two black holes in a tight orbit around each other). These can be strong sources of gravitational waves, as well as high-energy transients detectable in astronomical surveys. This work will also support the principal investigator’s student-oriented research program in computational astrophysics and will contribute to the training of the next generation of US scientists in supercomputing and the use of artificial intelligence and machine learning techniques. Education and public outreach activities will take advantage of the proposing team's close ties to the Adler Planetarium in Chicago and the Dearborn Observatory on the Northwestern campus in Evanston. This project will address several key questions concerning the formation and dynamical evolution of compact objects