This book offers a practical playbook for integrating agentic workflows throughout the building lifecycle. It demonstrates how to use Retrieval-Augmented Generation (RAG) to ground AI in project-specific data and highlights the use of open-source tools like OpenClaw for secure, local automation.

Generative AI and multimodal models offer immense potential, yet the AEC industry struggles to move from “cool demos” to reliable, high-stakes production. This book bridges that gap by introducing a socio-technical framework that connects built-environment information ecologies with LLMs and Computer Vision.

From automated code compliance in design to real-time UAV analysis in operations, readers find step-by-step guides and validation templates to mitigate risks like hallucinations and prompt injection. Written for researchers, CTOs, and practitioners, this book is essential for building trustworthy, efficient, and sustainable AI systems in the 2026 landscape.

Chi-Yun Liu is an assistant professor in the Department of Civil Engineering at National Taipei University of Technology (Taipei Tech) in Taiwan. He earned his Ph.D. from National Taiwan University of Science and Technology and later served as a postdoctoral researcher at Arizona State University. His research explores how generative and multimodal AI can improve the way construction projects and infrastructure assets are planned and managed. He is particularly interested in agentic systems that integrate large language models with field evidence while keeping outputs traceable and decisions subject to professional judgment. His work also examines field robotics and human–robot collaboration in contexts where automation should improve performance without weakening accountability. In Agentic Workflows for the Built Environment, he presents a human-centered framework for bounded autonomy across the asset lifecycle.