Answers are grounded in this paper and cite their sources.
Highlights the effectiveness of an AI learning assistant for human-building education, suggesting improved interdisciplinary skill development.
Human-Building Interaction (HBI) has emerged as a key domain within building science - typically encompassing civil/architectural engineering and architecture - in the past decade, as its potential to improve building performance by integrating human dynamics into the control of buildings. Educating students in HBI presents unique challenges due to its interdisciplinary nature. It requires multiple skills, including programming (e.g., Arduino and Python), wiring circuits with sensors, and understanding diverse domains, including thermal comfort, lighting, indoor air quality and more. We aim to address this challenge by developing an artificial intelligence (AI) learning assistant, incorporating web-based multi-agent frameworks for HBI education. In this study, we highlight this assistant's expandability, accessibility, and enhanced usefulness compared to general-purpose large language models (LLMs). Specifically, this assistant is supported through both LangChain and LangGraph, to combine the flexible integration of diverse LLM capabilities with the structured orchestration of complex workflows, thereby enabling adaptive and educational interactions for HBI education. We showed its usefulness in supporting Arduino programming by comparing its response with those generated by a generalpurpose LLM model using systematic evaluation criteria. This research and development effort will contribute to advancing educational innovation in HBI as well as interdisciplinary subjects in architecture, engineering, and construction by lowering technical barriers for students, fostering interdisciplinary skill development, and creating a scalable framework where AI supports experiential learning and empowers learners to transition from passive knowledge absorption to active problem-solving and innovation.
No takes yet. Share an insight, caveat, or question.
Jung et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: