Designing Reusable LLM-Enhanced Assignments: A Quality-Oriented Framework for Software Engineering Education

Authors:

Abstract
Large Language Models (LLMs) are increasingly prevalent in software engineering (SE) practice, yet their integration into education remains improvised and lacks theoretical grounding. This paper presents a quality-oriented framework for LLM integration in programming curricula, drawing on Cognitive Load Theory and Constructivism. We demonstrate its utility by using an LLM (Claude Sonnet 3.7) to redesign an Operating Systems assignment, followed by expert evaluation. Results show that LLMs can serve as tools for curriculum design and aids for student learning. For instructors, LLMs streamline the redesign of assignments while maintaining quality. For learners, they provide structured guidance and promote metacognitive development. We position LLMs as reusable educational components that can improve learning outcomes for students and streamline material design for instructors, while preparing both for professional environments where Artificial Intelligence (AI) collaboration is increasingly expected.