The collisions between neutron stars and between neutron stars and black holes are among the most energetic phenomena in the Universe. These events can be studied using gravitational-wave observatories, such as NSF's Laser Interferometer Gravitational-Wave Observatory (LIGO), and ground-based and space-based observatories, such as NSF's Vera Rubin Observatory and NASA's James Webb Space Telescope. Through these cosmic collisions it is possible to study, among others, the nature of matter at supernuclear densities, testing quantum chromo dynamics in a regime that cannot be probed on Earth, the nuclear physics involved in the formation of rare-earth elements, gold, and alkali metals, and the outcome of the evolution of massive stars. This project aims to develop the theoretical framework necessary to interpret upcoming observations and advance research in these areas. In particular, the team will construct a large template of synthetic neutron star merger observations, leveraging new national computing resources to perform sophisticated general-relativistic simulations. These data will then be used to develop statistical and artificial intelligence models that can be used to interpret real observations. This project will train one US graduate student and three US undergraduate students at the interface between high-performance computing, computational fluid dynamics, and machine learning, thus strengthening the US STEM workforce. This project will also develop new simulation