Scalable Methods for Customized Digital Learning is a toolkit that provides education application developers with algorithms for assessment of student proficiency. Precise assessment of students’ strengths and weaknesses makes it possible to customize learning modules, find the right balance between easy and difficult materials, and substantially improve learning outcomes. These algorithms will make this type of personalization possible, and the accompanying toolkit includes notebooks that guide developers through comparing methods, evaluating performance, and creating teacher dashboards.