What This Book Argues

This section is a compact statement of the book's argument, intended for orientation and later reference. This book makes six central claims. Together they explain how commodity intelligence changes the economics of engineering, why scale remains a problem, and how MAGE responds.

MAGE is a framework for governing agentic engineering work. This book develops that framework through software engineering, its first worked domain and the source of its evidence. The six claims below are therefore stated primarily in software-engineering terms; Chapter 6 abstracts their mechanisms, and Chapter 7 asks what transfers beyond software.

  1. Commodity intelligence changes the economics of software engineering. You can now purchase a kind of intelligence. The resulting implementation capacity is becoming abundant relative to engineering judgment. As implementation gets cheaper, engineering effort shifts toward what remains scarce: deciding what to build, representing the system clearly enough to reason about it, producing evidence, and making requirements enforceable by the engineering environment.
  1. Scale creates a reasoning problem. Large software systems exceed the reasoning horizon of humans; agents have the same problem. Software engineering has always answered scale with abstraction. Commodity intelligence does not remove that need. It makes the representations that guide the work more important.
  1. Purposeful models can reduce the cost of engineering reasoning. Engineering models capture the knowledge and intent needed for decisions. Software engineers have always used such abstractions to reason about architecture and other system properties. As commodity intelligence lowers the cost of deriving, maintaining, and using such representations, more engineering knowledge can be carried forward rather than held implicit or reconstructed from implementation.
  1. Enforcing obligations makes autonomous implementation governable. When agents perform more of the implementation, deciding what the system must do becomes increasingly separate from writing the code that does it. Important engineering decisions therefore cannot live only in instructions to an agent or in a person's head. Engineers can enforce these obligations through checks and controls that constrain work, distinguishing choices an agent is free to make from decisions engineers have already made. Engineers can then give agents substantial freedom in how they build the system while retaining control over the properties that matter.
  1. Governance conversion turns recurring judgment into durable engineering structure. When a failure exposes missing knowledge or an unenforced obligation, encode the lesson into a model, procedure, or mechanism that future work can inherit. Durable structure becomes engineering capital when later work keeps benefiting from it.
  1. Engineering work will reorganize around what remains scarce. As implementation becomes cheaper, engineering effort will move toward representation, evidence, governance, coordination, and judgment. Agents may perform increasing portions of that work as well. The enduring boundary will not be a particular task or level of abstraction, but responsibility for deciding what matters, what evidence is sufficient, which obligations should be enforced, and what tradeoffs remain acceptable.

In operation, MAGE has two principles and a conversion loop. Modeling makes consequential knowledge and intent explicit, while Alignment makes selected obligations enforceable. As engineering work progresses, gaps are inevitable. Governance conversion turns recurring human judgment into explicit engineering structure that future work can inherit. Figure 0.2-1 connects these responses to the economic change, the reasoning problem, and the predicted reorganization of engineering work.

The six claims, connected Claims one and two motivate two parallel principles: Modeling and Alignment. Both guide and govern engineering work. Recurring failures and judgment lead to governance conversion; feedback adapts models and controls for future work. Engineering capital depends on later benefit. Claim six is a separate prediction about the reorganization of engineering work. 1 · Commodity intelligence changes the economics of software engineering Implementation capacity becomes abundant relative to engineering judgment. 2 · Scale creates a reasoning problem Large software systems exceed the reasoning horizon of humans and agents. 3 · Modeling Purposeful models can reduce the cost of engineering reasoning. Capture consequential knowledge and intent while leaving irrelevant choices open. 4 · Alignment Explicit obligations can make autonomous implementation governable. Constraints restrict; sensors observe; gates control admission. Governed engineering work Substantial implementation freedom; control over the properties that matter. 5 · Governance conversion Recurring judgment becomes durable structure. Future work inherits models and controls. Engineering capital when later work keeps benefiting. 6 · Predicted reorganization of engineering work Engineering work will reorganize around what remains scarce. Representation · evidence · governance · coordination · judgment Responsibility for consequential decisions remains. motivate the two principles guide and govern recurring failures and judgment adapt models adapt controls
Figure 0.2-1. The six claims of MAGE. When engineers find themselves making the same judgment repeatedly, they can build it into a model or a control. Later work inherits that judgment, which can lower the cost of meeting the same obligation again.
© James C. Davis, 2026–present