AI Coding Maturity
Framework describing progressive levels of sophistication in how developers use AI for software development.
Also known as: AI Coding Maturity Levels, AI Coding Maturity Model
Category: AI
Tags: ai, software-engineering, frameworks, development-tools
Explanation
AI Coding Maturity is an 8-stage maturity model for AI-assisted software development, introduced by Steve Yegge in his "Welcome to Gas Town" article (January 2026). It describes the progression from minimal AI usage to fully orchestrated multi-agent workflows.
## The 8 stages
**Stage 1: Zero or Near-Zero AI** - Maybe code completions, sometimes asking chat questions.
**Stage 2: Coding agent in IDE, permissions on** - A narrow coding agent in a sidebar asks permission to run tools.
**Stage 3: Agent in IDE, YOLO mode** - Trust goes up. Permissions are turned off, the agent gets wider latitude.
**Stage 4: In IDE, wide agent** - The agent gradually grows to fill the screen. Code is just for reviewing diffs.
**Stage 5: CLI, single agent, YOLO** - Diffs scroll by. You may or may not look at them.
**Stage 6: CLI, multi-agent, YOLO** - You regularly use 3 to 5 parallel instances. You are very fast.
**Stage 7: 10+ agents, hand-managed** - Pushing the limits of hand-management. Orchestration tools become necessary.
**Stage 8: Building your own orchestrator** - You are on the frontier, automating your workflow.
## Key patterns in the progression
- **Trust escalation**: Each stage involves giving the agent more autonomy
- **IDE to CLI**: The transition from IDE-based to CLI-based marks a major shift in workflow
- **Single to multi-agent**: Going from one agent to many fundamentally changes the bottleneck from coding to design and planning
- **Design becomes the constraint**: At stages 7-8, the agent churns through implementation so fast that imagining what to build and planning the details becomes the limiting factor
Related Concepts
← Back to all concepts