This project aims to develop a next-generation AI-driven document reasoning system that integrates Retrieval-Augmented Generation (RAG) and Context-Augmented Generation (CAG) to automate complex information synthesis and decision support in construction and infrastructure projects. By combining advanced natural language processing (NLP), domain-specific knowledge graphs, and real-time data integration from Building Information Modelling (BIM), schedules, and site records, the system will enable intelligent interpretation, linking, and reasoning across diverse technical sources. The research addresses fundamental scientific challenges in multimodal data alignment, contextual knowledge retrieval, and AI explainability within high-stakes environments. It advances methodologies for human–AI collaboration and trust calibration, ensuring that decision-making in construction remains transparent and auditable. The project introduces an innovative AI architecture capable of bridging natural language and structured engineering data, enabling explainable insights and actionable recommendations. Aligned with the Dubai Future Foundation’s (DFF) RDI Grant Program, the project supports priority sectors in artificial intelligence, digital construction, and advanced manufacturing. It directly contributes to Dubai’s strategic vision of becoming a global hub for innovation and smart infrastructure, enhancing automation, efficiency, and data-driven governance. Expected outcomes include a prototype AI reasoning engine, a construction-aware knowledge base, and a web-based decision-support platform for engineers and project managers. Through field validation, benchmark datasets, and academic dissemination, the project will establish a new standard for intelligent document analysis in the built environment. Ultimately, this research will foster local AI capacity, strengthen Dubai’s RDI ecosystem, and accelerate the digital transformation of the construction sector through scalable, explainable, and context-aware AI technologies.