The exponential growth and increasing intricacy of Machine Learning (ML) technologies makes it one of the most critical sectors in higher education when it comes to pedagogical concerns. The difficulty of patchwork curricula in accommodating the varying learning styles and the disparate foundational knowledge of students can be overwhelming. We put forward an AI-Driven Adaptive Learning System for Personalised Machine Learning Education. Whenever an instructional activity is completed, Bayesian Knowledge Tracing (BKT), the student modelling engine, determine the conceptual understanding of students in real-time. The adaptive sequencing engine utilises this knowledge and the mastery-cognitive gap theory to close the gap and optimize mastery, and constructs personalised learning pathways and sequencing for the student in the mastery of various components. This includes conceptual articles, video lectures, and coding exercises. The frontend has been completed using React, which implements a responsive, interactive user interface including a browser-based code editor and a “Knowledge Map.” Backed by Python's Flask microservice, the backend runs the BKYT model, serves content, manages user data in a PostgreSQL database, and provides dynamic changes as opposed to a static content repository. Replacing static content repositories with dynamic adaptable tutoring models enhances the quality of the tutoring provided.