Although Generative Artificial Intelligence (GAI) presents vast opportunities and potential within the educational domain, the optimal integration of educational theory with GAI to enhance academic performance and problem-solving skills remains an area that requires further exploration. Therefore, this study aims to develop an integrating GAI with self-explanations scaffolding platform to enhance students’ learning performance and problem-solving skills through completing scaffolding tasks. Furthermore, this study evaluates students’ perceived ease of use and perceived usefulness of the system based on the Technology Acceptance Model. The findings indicate that students in the experimental group demonstrated a significant improvement in learning performance and problem-solving skills after utilizing the platform proposed by this study. Additionally, the experimental group exhibited positive technology acceptance toward the platform. Finally, results from the semi-structured interviews suggest that learning programming through the self-explanations scaffolding platform can mitigate learning setbacks, provide cognitive guidance, strengthen conceptual connections, facilitate knowledge construction, and enhance application skills.