Leveraging Artificial Intelligence (AI) in education has become one of the fastest-growing research areas in computer science. In recognition of its transformative potential, the National Science Foundation (NSF) has funded 6 out of 19 AI Institutes in the past four years, specifically dedicated to advancing AI-driven educational technologies. These initiatives have accelerated innovation across various learning environments, powered by EdTech tools such as Learning Management Systems (LMS), Intelligent Tutoring Systems (ITS), and other interactive digital platforms. The widespread adoption of these tools has enhanced both the teaching effectiveness and students' learning experience. However, these advancements also raise critical research questions that must be addressed: (1) What is the short-term and long-term impact of the EdTech tools? (2) How can we model, reconstruct, and predict user interaction with the system? (3) How can AI be harnessed to enable advanced features for EdTech tools?

To tackle these critical questions, my research employs two foundational methodologies: User Modeling and AI Adoption. User modeling within educational technologies (EdTech) enables the extraction of actionable insights and the development of a data-driven understanding of learning behaviors. This process reveals emerging needs and challenges encountered by students and educators across diverse learning contexts. Key techniques from Data Science and Human-Computer Interaction (HCI), including User Simulation, Data Mining, and Sequence Modeling, are integral to this approach. Building on these insights, my research aims to enhance or redesign EdTech tools by incorporating advanced AI capabilities (e.g., Predictive Modeling, Generative AI) through Software Engineering (SE) Integration.