Simulations of human movement that predict how a person moves based on a neuro-musculoskeletal model can be used to enhance our understanding of the interaction between motor control and musculoskeletal dynamics, and ultimately to design optimal treatments for neuro-musculoskeletal disorders. I will introduce our framework for rapid predictive simulations and demonstrate that our simulations capture the mechanics, energetics, and control (when uncertainty is accounted for) of a range of human gaits. I will share insights in neuromechanics of human movement gained from combining experimental data and simulations, and illustrate the framework’s potential to support treatment selection and design.