Generative AI in Software Engineering Education: A Study on Competency and Perception

Abstract

Software Engineering (SE) education struggles to align graduate competencies with industry needs, a challenge compounded by Generative AI (GenAI). This study’s purpose is to explore how structured GenAI utilization can enhance student competencies and learning perceptions. This exploratory case study used a mixed-method design involving four students in a controlled experiment. Data was collected via quantitative (SWEBOK-based artifact assessments, questionnaires) and qualitative narrative reflections. Results show the group using AI produced higher-quality artifacts and reported a more efficient, positive experience. A key finding is the transformation of work processes, shifting the primary challenge from manual execution to the critical evaluation of AI outputs. Due to its small scale, the findings are preliminary but suggest GenAI can enhance competencies. This requires an educational shift toward human-AI collaboration skills like critical evaluation and prompt engineering. Further validation with larger, more comprehensive studies is essential to generalize these initial insights.