Computational models of cognitive development could be used to optimize educational interventions by predicting their outcomes. Similar to how computational drug screening is used in medicine to identify the most promising drug candidates for clinical testing, computational models could screen millions of potential educational interventions to identify the most effective lead interventions for empirical testing. This approach could explore parts of the intervention space that would be unethical or impractical to test directly on children and help identify optimal trajectories for student development. However, questions remain about how accurately these models map onto real child development and learning.