Abstract: Functionality is imparted by interfaces, and an atomic-scale understanding of interfacial chemical physics is needed to address fundamental questions in energy storage and conversion, sensing, geochemistry, and even the origins of life. These areas are unified by challenges presented by the coupling between solid-state electronic processes and solution-phase chemical dynamics, from electrode-electrolyte interfaces in batteries to liquid-semiconductor interfaces in sensing and catalysis. In this talk, I will discuss our ongoing work focused on understanding and modeling electronic effects in liquids, solids, and liquid-solid interfaces. I will first discuss the role of long-range interactions in liquids and how these can be included in machine learning-based neural network models, including electronic polarization. The resulting neural network model is transferable to systems in the presence of applied fields and at interfaces despite being developed only for the bulk, addressing the problem of transferability in machine learning models. Then, I will discuss dynamic fluctuations of lone pair electrons in crystalline solids. Lone pair electrons can exhibit dynamic orientational disorder, similar to plastic phases of molecular solids, and the resulting electronic rotations govern important processes in materials. I will show that lone pair rotational dynamics play a key role in solid-state ion conduction and in determining structural, vibrational, and dielectric properties of metal halide perovskites used for photochemical applications. Finally, I will discuss a new approach for modeling electron and hole quasiparticles in two-dimensional semiconductors at interfaces with liquids. We use this approach to quantify the effects of interfacial liquids on screening of electron-defect interactions, effects of liquid screening on electronic transport, and the effects of defect-induced trap states on interfacial properties relevant to catalysis.