Bradley Jolicoeur
Boards are telling engineering leaders to deliver two, three, even ten times more software with AI. But when output surges faster than review, decision-making, and customer learning, the result is not transformation. It is congestion, rework, incidents, and a growing loss of trust.At Pinnacore, engineering director Marcus Vance is living that reality. Under relentless executive pressure to prove AI can triple delivery, he watches pull requests pile up, product priorities thrash, and production stability erode. The company celebrates speed while the system that turns code into customer value starts to fracture. Every team is working harder. Every dashboard says more is happening. Yet the business feels less in control by the week.Then Mac, an unconventional mentor, cuts through the noise with a blunt question: 'If AI can build a feature in two hours, why did it take you six weeks to learn nobody wanted it?' In one line, he exposes the real bottleneck. AI has made building faster, but it has not solved the harder problem of knowing what is worth building, when work is truly ready, or where flow is actually breaking down.As Pinnacore edges toward crisis, Mac introduces Marcus to the Theory of Constraints. What follows is not a promise of easy acceleration, but a new way to see the system: speed is not flow, local efficiency is not system performance, and more output is not more value. To steady the company, Marcus and his team must discover where work truly stalls, how AI can amplify the wrong constraint, and what it takes to restore focus, feedback, and trust across the whole organization.In the tradition of The Goal and The Phoenix Project, The Constraint Loop is a business novel for the age of AI-assisted software development, written for engineering leaders, DevOps practitioners, TOC-Lean readers, and fans of business novels that turn operational insight into compelling fiction.
