Simon Willison, a prominent open-source developer, has launched "Agentic Engineering Patterns" to codify software development in an era of autonomous AI agents. Unlike "vibe coding"—where users prompt AI without deep technical oversight—this initiative defines Agentic Engineering as a disciplined approach for professionals. It focuses on using tools that can autonomously write, run, and iterate on code until a specific goal is achieved without constant human guidance.

The project is structured as an "evergreen" digital guide, drawing inspiration from the classic software design patterns that revolutionized object-oriented programming in the 1990s. The first chapters address fundamental shifts in the industry, specifically how the cost of generating initial code drafts has plummeted. This shift forces engineers to move their primary focus from the manual act of writing code to the high-level tasks of reviewing, testing, and system architecting.

A central strategy highlighted in the guide is Red/Green Test-Driven Development (TDD). By writing tests before the AI generates code, developers provide a clear success criterion that the agent uses to verify its work independently. This iterative loop ensures that the resulting software is both reliable and succinct. These patterns represent a broader transition where AI moves from a simple completion assistant to a sophisticated collaborative partner in complex engineering workflows.

Simon Willison, a prominent open-source developer, has launched "Agentic Engineering Patterns" to codify software development in an era of autonomous AI agents. Unlike "vibe coding"—where users prompt AI without deep technical oversight—this initiative defines Agentic Engineering as a disciplined approach for professionals. It focuses on using tools that can autonomously write, run, and iterate on code until a specific goal is achieved without constant human guidance.

The project is structured as an "evergreen" digital guide, drawing inspiration from the classic software design patterns that revolutionized object-oriented programming in the 1990s. The first chapters address fundamental shifts in the industry, specifically how the cost of generating initial code drafts has plummeted. This shift forces engineers to move their primary focus from the manual act of writing code to the high-level tasks of reviewing, testing, and system architecting.

A central strategy highlighted in the guide is Red/Green Test-Driven Development (TDD). By writing tests before the AI generates code, developers provide a clear success criterion that the agent uses to verify its work independently. This iterative loop ensures that the resulting software is both reliable and succinct. These patterns represent a broader transition where AI moves from a simple completion assistant to a sophisticated collaborative partner in complex engineering workflows.