Basic Programming is one of the fundamental courses in computer science education, yet many beginner students experience difficulties in understanding programming concepts due to limited feedback, lack of personalized guidance, and challenges in debugging code. Conventional learning methods often rely on lectures and static learning materials, which are often insufficient to support interactive and adaptive learning processes. This study aims to develop an AI-Assisted Learning Media for Basic Programming Courses using a web-based platform integrated with a Large Language Model (LLM) and supporting machine learning techniques. The research employed a Research and Development (R&D) approach using the Waterfall model, including requirement analysis, system design, implementation, testing, and evaluation stages. The developed system integrates programming materials, coding exercises, intelligent tutoring support, debugging assistance, and learning progress monitoring into a single platform. The results show that the LLM-based tutoring model achieved better performance than the XGBoost model, with an accuracy of 0.91, precision of 0.89, recall of 0.90, and F1-score of 0.89. These findings indicate that AI-assisted learning media can improve programming education by providing adaptive learning support, real-time tutoring, and effective feedback for beginner students.