Vibe Coding for Simulation Development in Engineering Education

ABSTRACT Simulation‐based learning is widely adopted in engineering education because it enables learners to explore complex engineering systems through interactive and self‐paced learning. However, developing educational simulations traditionally requires both domain expertise and software development skills. Recent advances in generative artificial intelligence (GenAI) have introduced vibe coding , in which software is created primarily through natural‐language prompting, potentially lowering the technical barrier to educational simulation development. While existing research mainly focuses on AI‐assisted programming from either a practical or a learning perspective, this study presents a preliminary investigation of vibe coding for developing educational simulations through two undergraduate engineering case studies: one in automotive engineering and the other in communication engineering. A practical workflow for simulation development is proposed, and the development process is analyzed to identify both its benefits and challenges. The findings demonstrate that functional web‐based educational simulations can be developed efficiently without requiring extensive programming or prompt engineering expertise. However, the study also shows that instructors remain responsible for validating engineering concepts, verifying simulation behavior, and diagnosing conceptual implementation errors that could not be resolved through iterative prompting alone. Furthermore, the developed simulations were found to align with established simulation design principles reported in the literature. The findings suggest that AI‐assisted vibe coding can substantially reduce the technical barriers to educational simulation development while preserving the central role of educators in instructional design and domain‐specific validation.