Undergraduate course scheduling is a complex task that significantly impacts students’ academic success and satisfaction. We present a novel automated scheduling system that combines sequence-to-sequence (Seq2Seq) models, BERT-based models, and rule-based systems to generate personalized academic schedules. Our system considers user-specified criteria such as course preferences and graduation timelines while ensuring compliance with academic constraints like prerequisites and unit limits. We tested the BERT model for accurate extraction of scheduling variables and the Seq2Seq model for producing valid schedules that adhere to all academic rules. The results demonstrate that integrating advanced machine learning models with rule-based systems can effectively streamline the scheduling process and improve academic planning.