Biomedicine currently faces many large challenges, many of which can be directly related to the Crisis of Reproducibility. The use of multi-scale modeling, specifically agent-based models, can help to address these challenges by serving as proxy systems upon which high throughput experimentation can be performed.  This pipeline can be significantly augmented through the use of high-performance computing and machine learning.  In this talk, I present two specific use cases of machine learning and high-performance computing to enhance the utility of dynamic computational modeling: the exploration of a large parameter space to constrain model behavior, and the development of control strategies to provide a tractable path towards Precision Medicine (i.e., right drug, right patient, right time).