The convergence of agentic artificial intelligence (AI), physics-informed machine learning (PIML), and advanced computational mechanics is opening new frontiers in engineering design and manufacturing. Agentic AI systems move beyond passive prediction by autonomously planning, reasoning, and adapting across design–simulation–fabrication loops, enabling faster and more reliable solutions to complex, high-dimensional optimization problems. In parallel, PIML embeds governing physical laws into AI models, delivering robust generalization even with sparse, biased, or noisy data. This integration enables next-generation engineering pipelines that are data-efficient, physics-consistent, and human-centered. These methods have already shown impact in areas such as materials discovery, topology and shape optimization, process control, and multi-physics system design. However, significant challenges remain in scaling these approaches to industrial environments, bridging computational innovations with manufacturing realities, and ensuring trustworthiness and interpretability in deployment. The aim of this minisymposium is to bring together researchers and practitioners working at the intersection of computational mechanics, AI, and manufacturing. We seek to highlight recent methodological advances, share successful application cases, and discuss remaining barriers to adoption. By fostering a cross-disciplinary dialogue, the symposium will explore pathways toward scalable, industry-ready AI-driven design frameworks capable of transforming manufacturing ecosystems. Topics include, but are not limited to: Agentic AI for autonomous design–manufacturing workflows Physics-informed neural networks (PINNs) and neural operators Multi-agent and LLM-based collaborative optimization Generative AI for inverse design of materials and processes AI-augmented topology optimization and multi-physics simulation