All-atom simulations provide a means to study processes such as biomolecular self-assembly in atomistic detail. The limits of the technique are in the computational cost of performing such calculations. Techniques such as implicit solvation and coarse-graining are used to accelerate simulations. Unfortunately, the most widely used implicit solvent models are either too simplistic and unreliable (such as surface area calculations) or too computationally expensive (such as RISM calculations). We are developing an implicit solvent model that is computationally efficient to be used in simulation while also being physically accurate. This work thus spans studying the theoretical chemistry of solvation while also the practical aspects of designing efficient algorithms to capture these physics. Recently we published our work on studying hydrophobic solutes with an Implicit Solvent using the SuperPosition Approximation (IS-SPA) yielding computational speed-ups of several hundreds. Currently we are looking to use the insights from IS-SPA to better understand hydrophobic interactions and also expand IS-SPA to capture dynamics and simulate charged solutes. Ultimately this model will allow for simulating unprecedented size- and time-scales of the self-assembling biomolecules being studied in the McCullagh Group.