Jason and I discuss what breaks when you try to use agentic AI in complex engineering systems, where today’s agentic AI tools tend to fall short for engineering teams, and why AI must integrate with simulation, verification, and deterministic tools to produce repeatable, traceable evidence. Links for August 7, 2026Guide to Agentic AI with MATLAB and Simulink (ebook)Can Agentic AI Develop Embedded Systems with Model-Based Design (14-minute video)Innovative Air Search and Rescue (2-minute video)Click here to check out the Fish Fry Archive. Jason and I discuss what breaks when you try to use agentic AI in complex engineering systems, where today’s agentic AI tools tend to fall short for engineering teams, and why grounded AI—the kind that actually works within your existing engineering workflows—is the only way to go. Jason Ghidella:I think the difference is that engineering work is tied to system behavior, and engineering evidence has to be repeatable. Grounded AI also means respecting the boundary between what AI should do and what deterministic engineering tools should do.