A New Way to Teach Programming using Generative AI and Modular Mastery Threshold Pedagogy (MMTP)

This paper presents the Modular Mastery Threshold Pedagogy (MMTP), a novel educational framework that synergistically combines three established approaches: mastery-based learning with its focus on complete skill acquisition, competency-based education that emphasizes demonstrable abilities, and modular learning that structures content into sequential building blocks. By organizing curriculum into competency-based modules with clear mastery thresholds, this approach emphasizes skill acquisition over point accumulation while maintaining traditional letter grades. The system employs pass/fail rubrics with 100% mastery requirements, authentic assessment tasks, and opportunities for resubmission, fostering a growth mindset. Students progress through tiered modules (basic to mastery level) that correspond to traditional letter grades, providing both flexibility and academic rigor. In this paper, we detail our implementation of this framework in an introductory Python programming course featuring seven progressive modules that transform students from complete beginners to developers capable of building full-stack web applications with generative AI integration. Our implementation experiences reveal significant benefits including enhanced student motivation through clear progression pathways, improved assessment transparency, and deeper learning outcomes. The paper addresses implementation challenges including student resistance, instructor workload, and rubric alignment, while providing practical mitigation strategies for each. This approach not only meets contemporary educational demands for skills-focused instruction but also maintains institutional compatibility by integrating with existing academic frameworks, offering educators a pragmatic pathway to emphasize competency development in higher education.