2025 ASEE Annual Conference & Exposition

Higher education continues to respond to the challenges and opportunities presented by artificial intelligence (AI) and large language models (LLM) such as chatGPT. In a prior work we introduced a chatbot that used AI and LLM to recruit prospective students, assist current students with academic advising (course selection, changing majors) and student affairs (directing to university resources regarding the campus community, housing and dining, student organizations, mental health and more. Towards the promotion of student success initiatives we report in this work our formulation of course specific teaching assistants for engineering and computer science. Through inquiry-aided interaction, the Barkplug chatbot helps students assemble prompts that aid them in the most challenging topics within core courses, and allow them to understand methodologies for solving homework problems and prepare for midterm exams. This paper will specifically discuss algorithm development and outline some case studies relevant to improving persistence, retention and graduation. Furthermore, while LLMs are generally considered a good tool for inferences and explanations, they suffer in calculations and involved engineering applications. This work highlights steps that would allow LLMs to work as reliable teaching assistants, serving both science and engineering oriented students.