Congenital heart defects affect approximately one in every hundred births and are the leading cause of infant mortality in the United States. Despite successes in surgical treatment, suboptimal outcomes remain common. Surgical treatment of complex, rare congenital heart valve defects typically follows an empirical, retrospective, “guess and check” approach. Further, children with congenital valve defects are a heterogeneous patient population with limited numbers, so clinical trials are difficult to conduct. Thus, there is an unmet clinical need for engineering tools for design and analysis of congenital valve repair. Simulation-guided design tools provide a flexible, controllable and efficient means to predict optimal surgical repairs and fill this clinical need. This talk will present new methods for fluid-structure interaction simulations of heart valves and the application of these methods to congenital heart disease. I will present a novel, nearly first-principles method for model generation called design-based elasticity. In this method, a system of partial differential equations representing the mechanical equilibrium of the valve under pressure is derived. The solution of these equations, via tuning parameters and boundary conditions, is designed to represent the predicted loaded configuration of the valve. A full model is then constructed from the loaded configuration. When simulating their coupled dynamics with blood, these models are highly effective, producing realistic flows under physiological pressures over multiple cardiac cycles. Results compare favorably with experimental data and show that hemodynamics change drastically with congenitally diseased valve phenotypes. Finally, I will discuss simulation-guided design of surgical bicuspidization of the aortic valve, a repair technique for severe congenital aortic pathology, and preliminary results on in vitro validation and clinical translation to surgical practice.