2025 ASEE Annual Conference & Exposition

The rapid development of large language models (LLMs) in recent years has ushered in a new wave of transformation in our day-to-day lives. This advancement significantly impacts the field of academia, where LLMs are transforming the way students and educators engage with information. We can compare this technological leap to the revolution of search engines like Google, which fundamentally transformed how we access information. In this paper, we will present an AI-powered platform that aims to improve the educational experience for both students and instructors in university settings. While there are several AI tutor models already available, our platform will be designed to go beyond merely answering student inquiries or supporting basic learning tasks. We will build a modern, AI-infused education system that elevates the experience for both students and educators. Unlike traditional AI models, which focus primarily on addressing doubts, our platform will function as a 24/7 AI assistant that adapts to each student’s unique learning journey. We will explore Deep Knowledge Tracing (DKT) and other potential methods to implement knowledge tracing, a method that tracks a student's understanding over time by analyzing their interactions with learning material. Our platform ensures that each student receives personalized support tailored to their current level of understanding, avoiding rote solutions and promoting deeper learning. Importantly, while safeguarding student privacy, the platform provides instructors with valuable summary reports that highlight common areas of confusion across their class. This allows educators to adapt their teaching strategies and prioritize challenging topics more effectively.

Moreover, our system will seamlessly integrate with instructors' workflows, offering assistance in lesson planning and generating insights based on student performance and queries. This will enable teachers to focus on what they do best—teaching, while the platform handles data-driven insights and lesson optimization, creating a truly comprehensive learning environment that benefits both students and educators. The platform will be designed to integrate seamlessly with existing learning management systems (LMS) like Blackboard and Moodle via APIs and Learning Tools Interoperability (LTI), ensuring an easy and efficient adoption process for institutions. This integration allows the platform to work within the existing academic infrastructure, making it possible for students and educators to access its features without disrupting their current workflows. Additionally, we will also employ explainable AI (XAI) such as SHAP, ensuring transparency in decision-making and helping identify and address any biases, fostering trust and fairness for students. This platform will bridge the gap between learning and teaching through intelligent, adaptive, and explainable systems, contributing to advancements in educational technologies.