I'm a Civil Engineering Ph.D. candidate at Stevens Institute of Technology, working in the Smart Infrastructure Lab. My research develops AI-enabled structural engineering systems across the design and operation lifecycle, combining language-model agents, engineering knowledge, physics-guided models, digital twins, and structural health monitoring. I aim to make structural design, monitoring, and infrastructure decision-making more automated, interpretable, and verifiable. My work has appeared in Automation in Construction, Advanced Engineering Informatics, and Engineering Structures.
Download CVMy research asks how AI can support structural engineering across the full design–operation lifecycle while remaining grounded in mechanics and engineering knowledge. On the design side, I develop multi-agent language-model systems that interpret design codes, interact with structural software, and produce transparent, verifiable designs. On the operation side, I work on structural health monitoring and digital twins that combine sensing, data-driven methods, and physics-based models to estimate structural condition and support infrastructure decisions. Across both, I focus on connecting data, physics, engineering knowledge, and AI agents into reliable systems for structural design, monitoring, prediction, and decision-making.
Selected work across three connected themes: AI-enabled structural design, physics–data integration, and digital twins for structural health monitoring.
An AI design agent that integrates knowledge graphs and numerical models to automate the operation of structural software for engineering design workflows.
J. Chen, Y. Bao · 2026An LLM framework that decomposes structural design into specialized agents, reasons over design codes, and returns code-compliant, interpretable reinforced-concrete designs.
J. Chen, Y. Bao · 2025A data–physics hybrid framework in which an LLM coordinates finite-element analysis and genetic optimization for reliable, automated UHPC beam design.
J. Chen, Y. Bao · 2026A domain-specific LLM framework that uses retrieval-augmented generation over structural design codes to provide grounded, verifiable answers for engineering code interpretation.
J. Chen, Y. Bao · 2025Interactive prototypes that turn research ideas into working engineering systems, spanning code-aware structural assistants, research automation, and digital twins for structural monitoring.
Current digital-twin direction: linking sensing, structural response models, damage-state estimation, and AI-based prediction to support condition assessment and infrastructure decisions.
Notes and essays on AI-enabled structural design, digital twins, SHM, and intelligent infrastructure.
Expected to complete my Ph.D. in Summer 2027 and currently exploring postdoctoral opportunities for Summer/Fall 2027. I am also open to research collaborations and discussions in AI-enabled structural engineering, digital twins, and structural health monitoring.
[email protected]