In recent decades, simulation has become an indispensable element of materials design. The simulation of disordered materials presents unique challenges due to the large configuration space available and pathway dependence of material properties. In this dissertation, a wide variety of complex disordered materials are studied and designed. To begin, we explore disordered metamaterials, and create design rules for these materials to be tuned to show a negative Poisson's ratio. Such materials show promise in areas such as impact mitigation, filtration and as structural elements. These materials are anisotropic, however, meaning that their properties depend on the direction in which they are strained. We address this and other issues by introducing new optimization schemes which yield isotropic materials with Poisson's ratios approaching the lower mechanical limit. These optimized materials also show enhanced resistance to shear. We then transition to a study of highly stable glasses formed by a process of physical vapor deposition. Under particular conditions, physical vapor deposition of glass-forming molecules allows for films to be formed which are as stable as liquid-cooled glass films which have been aged for many years. However, a clear structural understanding of how these highly stable vapor-deposited glasses relate to their liquid-cooled counterparts is lacking. Here, we use a model glass-forming system to explore this problem. We then address two challenging topics in the area of organic photovoltaics: solution processing and solid-state film analysis. We introduce novel techniques to determine solubility in organic photovoltaic polymers, a necessary prerequisite for solution processing. The techniques allow for polymers to be rapidly designed and screened for solubility, and provides physical understanding of the origins of their solubility. Turning to solid-state film analysis, we introduce techniques to investigate interfacial properties of the ubiquitous bulk heterojunction, as well as microphase purity. Finally, we introduce a GPU-based molecular dynamics engine, DASH, which provides high performance and several new features within a flexible Python framework.