The background of this study focuses on the reform of the Ship Understanding Internship course in the context of engineering certification, aiming to improve teaching effectiveness and assessment quality through the introduction of an AI-intelligent tutoring system. In this study, an intelligent personalized tutoring system is constructed with the meta-learning algorithm and BERT model as the core, combined with ChatGPT. The system is able to answer students’ questions in real-time, provide personalized learning suggestions, and track students’ learning progress to provide teachers with accurate teaching feedback. In addition, this study explores the application of ChatGPT as an assessment tool in the process of a ship’s cognitive internship in order to reduce the burden on teachers and improve the efficiency and accuracy of evaluation. The results of the study show that the reform of the course “Ship Understanding Internship” based on AI intelligent tutoring system has achieved remarkable results. The intelligent tutoring system not only improves students’ learning efficiency and satisfaction, but also helps teachers optimize their teaching strategies through accurate teaching feedback. Meanwhile, the application of ChatGPT as an assessment tool effectively reduces the burden on teachers and improves the efficiency and objectivity of evaluation. The importance of this study lies in the fact that it provides new ideas and methods for the reform of the ship’s cognitive internship course and promotes the development of education in the direction of high quality and diversity. In addition, this study also provides useful reference and material for the reform of other practical courses. In the future, such systems are expected to be widely used in various types of practical teaching, providing students with more accurate and personalized learning support and promoting the sustainable development of the education industry.