Models Think, Systems Act: Rebuilding the Industrial Operating System

Three months into the agentic AI shift, manufacturing finds itself at the same inflection point a generation of CEOs faced at the dawn of digital: lean in now, or wait for someone else to prove it works. The cost of waiting just got steeper.

Enno de Boer, Senior Partner at McKinsey and co-founder of the World Economic Forum Global Lighthouse Network, returns to Augmented Ops to talk with Natan Linder, Tulip's co-founder and CEO, about why this is not the next wave of IoT, 3D printing, or even mobile. Enno argues AI is a general purpose technology on the order of electricity, the first one in roughly a century, and that most industrial organizations are still treating it like a tool layered on top of yesterday's workflows.

The result, he says, is Amdahl's Law in action: speed up one element by 20%, leave the system intact, and you get zero results. That, in his read, is why 2025 returned almost no measurable AI impact in industry, and why the new normal of a major supply-chain disruption every single year demands more than another wave of pilots and chat interfaces. The real work is to rebuild the operating system underneath, the AI system, the data system, and a new division of labor between humans and agents.

The conversation gets specific. Enno walks through John Deere's agentic material-availability and replanning system, a top-five electronics supply chain that hit a billion dollars of impact by collapsing planning cycles tenfold, and a coding-agent deployment in financial services where a night shift of agents drove 200x throughput while humans handled validation by day. He is blunt about what has decayed since the Toyota Production System era, why the 80/20 inversion of direct-to-indirect labor in modern factories has gone out of check, and why the future is distributed, hyper-local, and personalized down to an "order of one." McKinsey's Rewired playbook anchors the strategy lens: identify the two or three economic leverage points that drive 70% of the value, then build the systems to go after them.

For operations leaders, his ask is a single metric to optimize against: clock speed. Enno's ambition for the next decade is a 100x increase in manufacturing clock speed, and he is direct about what it will take to get there.

Models think. Systems act. Manufacturing's job, starting now, is to build the systems.

Watch the full episode on YouTube: https://youtu.be/KYGu7oOMh2A

Augmented Ops is a podcast for industrial leaders, citizen developers, shop floor operators, and anyone who cares about what the future of frontline operations will look like across industries. This show is presented by Tulip, the Frontline Operations Platform. You can find more from us at Tulip.co/podcast or by following the show on LinkedIn.