2026 ASEE Annual Conference & Exposition

Introductory programming courses often face challenges related to student engagement, scalability, and the transition from passive lectures to active, problem-based learning. While many students can follow coding syntax and isolated examples, they frequently struggle to apply concepts cohesively when faced with open-ended problem-solving tasks. This Work-in-Progress (WIP) paper presents the development of an AI-assisted instructional tool designed to bridge this gap by enabling real-time, interactive code challenges during class sessions.
The platform allows instructors to launch coding challenges aligned with lecture topics and instantly distribute them to students’ devices. Students can work individually or collaboratively in small groups, entering their code solutions directly into the system. The embedded AI assistant interacts with learners in natural language, providing conceptual explanations, contextual hints, and feedback on the logic of their approach, without directly revealing the correct answer. This conversational interaction simulates the presence of a personal teaching assistant, helping students diagnose misconceptions and reinforcing problem-solving autonomy.
Once challenges are completed, the system compiles responses and generates analytics that highlight common misconceptions, success rates, and patterns in problem-solving strategies. These analytics are then shared with both students and instructors, offering personalized and class-level feedback that can inform subsequent instruction. The design of this system aligns with current trends in computing education that emphasize formative assessment, metacognition, and the responsible integration of AI to enhance, rather than replace, human teaching.
This work also aims to prepare students for a future where AI systems increasingly assist in coding tasks. By engaging students in active problem-solving rather than passive code generation, the platform cultivates understanding of computational reasoning and helps learners internalize programming as a structured language for expressing logic, an essential skill for adapting to emerging AI-driven workflows.
A pilot implementation is planned across two large introductory programming course sections, where data on usability, learning outcomes, and student perceptions will be collected. The feedback from this trial will inform iterative improvements to the platform’s interface, AI tutoring behavior, and feedback mechanisms, with the long-term goal of enabling scalable adoption in diverse classroom contexts.
This WIP contributes to the growing body of research on AI-driven educational tools for engineering and computing education. It highlights an approach that combines pedagogy, software design, and responsible AI integration to support active learning and scalable instruction in large programming classrooms.