AI Coding Assistants Spur Push for Manufacturing‑Style Cost Controls

Authors Atul Parte and Sudarshan Rajput recount a two‑decade‑old question about applying real‑world cost accounting to software projects. Their professor dismissed it as impossible, noting that software cannot be fully described before it is built. Today, with AI‑driven coding assistants—often likened to a Formula One car for developers—teams experience divergent outcomes. Some see rapid delivery and soaring velocity, while others end up with chaotic codebases that are hard to review, predict, or maintain.

The authors argue that the technology itself is not the problem; rather, the lack of a disciplined production system, akin to manufacturing’s tolerance‑driven processes, leads to structural chaos. They propose a "Manufacturing Workbench Model" for AI‑native software development lifecycles, introducing manufacturing‑style costing, governance, and tolerance controls to keep AI‑generated code coherent and maintainable.