chapter one

1 Introduction to Intent-Driven Development

 

This chapter covers

  • What does intent mean in AI coding
  • Why intent articulation is the new bottleneck
  • What Intent-Driven Development is and how it works
  • Introduction to Spec-Driven Development and Loops

The productivity boost from modern coding agents is real, but it has exposed a deeper problem. As models grow more capable and execute larger stretches of work autonomously, our human-centric workflows, structured around the speed at which we build software, are being pushed to their limits. The bottleneck is shifting from writing code to articulating the intent that guides it.

Building a feature quickly with a coding agent is one thing. Fixing defects in that feature, adding capabilities to it, and sustaining the AI speed advantage over weeks and months of continued development is another. This requires systems that structure the information guiding coding agents, enabling highly aligned outcomes. Primitive techniques like vibe coding and ad hoc prompting cannot provide the kind of context that increasingly capable agents need to execute complex tasks independently.

Intent-Driven Development (IDD) is an AI-native engineering approach to software development that focuses on translating implicit human intent into usable, and consistent, language and systems for agents, enabling them to independently generate implementations that are highly aligned with that intent.

1.1 Understanding intent

1.1.1 Stating requirements without making intent explicit

1.1.2 What is “Intent”?

1.1.3 What "Intent" is not?

1.1.4 Intent articulation for AI

1.2 Intent articulation before AI coding

1.3 How intent-driven development works

1.4 Harnessing intent

1.4.1 Intent articulation mechanisms

1.5 Intent articulation is the new bottleneck

1.5.1 No mental model of the code

1.5.2 Fix intent articulation, not the implementation

1.6 When not to use Intent-Driven Development

1.7 What do you need for the rest of this book

1.8 Summary