This paper will explore the application of Physics-Informed Neural Networks (PINN) to the study of boiling heat transfer. Modern demands for efficiency require increased power density in electronic chips, battery cells, and power generation systems. While single-phase cooling is limited to a power density of around 100 W/cm2, the ceiling for multiphase is much higher. For this reason, there is significant motivation across a range of industries to obtain a greater understanding of the mechanisms involved in this extremely complex multiphysics process. This presentation will detail the development of a PINN methodology designed to resolve the boiling process without relying on empirical relationships commonly found in literature. This innovative process involves several novel case studies. One such study investigates rising bubbles within a convection cell, where the PINN algorithm achieved a peak positional error of 3.6%. Another novel application was the study of an evaporating vapour bubble in superheated liquid, where the PINN model could predict an unobserved configuration with a peak error of within 1.3% compared to a reference computational fluid dynamics (CFD) solution. Recently, the algorithm was applied to several film boiling studies. Here, the PINN methodology achieved an accuracy to within 5% of the reference CFD solution and strong qualitative results for both forward and inverse investigations. This study represents a pioneering effort in the development of PINNs for phase change by applying the current algorithm to boiling problems and provide guidance for developing a robust inverse training strategy.