The goal of this proposal is to support Computer Science (CS) education at the undergraduate level. The Qatar National Vision 2030 emphasizes excellence in education to foster the development of a knowledge-based society. However, current Computer Science degree programs in Qatar suffer from very low enrollment and high attrition rates, mostly due to low achievement of students in their freshman and sophomore years. Our goal is to support students in learning concepts and skills that are known to be "stumbling blocks" to achievement and retention in CS, such as linked lists and recursion. Towards this end, we propose to develop and rigorously evaluate Intelligent Tutoring Systems (ITSs) that can support student achievement in CS core courses. We will systematically investigate one-on-one tutoring as a main paradigm for effective teaching and learning. Within this paradigm, we will explore the effectiveness of modes of instruction (e.g., analogy and demonstration) that support learning from errors and uncertainty reduction. We will operationalize these modes, and uncover the contextual factors that trigger them. Then, we will (a) systematically embody these modes within a novel ITS, ChiQat-Tutor; (b) integrate ChiQat-Tutor into the undergraduate CS curricula of universities in Qatar and the United States; and (c) conduct systematic evaluations of the learning environment as ChiQat-Tutor is used in actual CS courses.