Terms  /  Human-Agent Relationship  /  Human-Code Divergence
07 · Human-Agent Relationship

Human-Code Divergence

The gap created between a user and their codebase when an agent mediates all development.

Human-Code Divergence is a new class of problem unique to agentic coding. The developer stops writing code directly and instead describes what they want to an agent that writes it. Over time, a gap opens: the developer knows less and less about how their own codebase works. They know what it does but not how. They can describe features but can't trace bugs. They own the product but not the implementation. This isn't inherently bad — abstraction is how all engineering progresses — but it becomes dangerous when the developer can't review, understand, or override the agent's work.

Example
A solo developer builds an entire application through an AI coding agent over six months. The app works well and has 40,000 lines of code. When a critical bug appears in production, the developer opens the codebase and realizes they can't find the relevant code. They don't recognize the patterns the agent used, can't follow the data flow, and don't understand why certain architectural decisions were made. They ask the agent to fix it, but the agent has lost the session context. The developer is locked out of their own codebase.

More in Human-Agent Relationship