Abstract: Modern nuclear theory seeks to understand how protons and neutrons self-organize to form atomic nuclei and the beta-equilibrated matter found in the interiors of neutron stars. A major challenge in this endeavor is solving the quantum many-body problem, which arises from the exponentially growing Hilbert space. In this talk, I will present a variational method that I have developed to overcome these limitations by employing artificial neural networks to compactly represent quantum many-body states. First, I will demonstrate how the shell structure of atomic nuclei naturally emerges from the fundamental interactions among protons and neutrons. Next, I will highlight how the flexibility of these wave functions enables us, for the first time, to uncover highly clustered forms of matter present in the interiors of neutron stars and how these clusters impact the star’s composition. Finally, I will discuss prospects for accessing the real-time dynamics of quantum many-body systems and their relevance to nuclear physics and the accelerator-based neutrino oscillation program.