Computational condensed matter physics and materials science is moving rapidly from merely computing material properties and explaining experimental results to designing new materials 'in silico'. Examples of large active research efforts include stronger alloys, new materials for higher capacity batteries, and more efficient solar cells. I will discuss the promise and challenge of computational materials design using the work of my research group on novel inorganic semiconductors, such as Bariumcopperfluoridechalcogenides (BaCuChF Ch={S,Se,Te}) and Copperphosphorselenide (CuP3Se4), as an example. These materials have potential applications in thin film solar cells and transparent electronics, which require the knowledge of optical, electronic, and interfacial properties, which in turn require increasingly complex calculations for ideal crystals, point defects, and extended interfaces. For (mainly optical) properties calculated from the ideal crystal structure of a material, we are in a position to make predictions with confidence. For (mainly electronic) properties, which cannot simply be determined from ideal crystal structures, predictions are less certain and much work remains to be done.