# AI Coding Dictionary

A glossary that explains AI-coding jargon in plain English, covering models, context windows, tool calls, failure modes such as hallucination, and patterns of working with coding agents.

The AI Coding Dictionary explains the vocabulary developers encounter when working with AI coding tools, translating jargon into plain English. It is organized into sections covering the model itself (parameters, training, inference, tokens), sessions and context windows, tools and environment, and common failure modes such as sycophancy and hallucination.

Later sections cover handoffs between sessions, memory and steering mechanisms such as AGENTS.md files and skills, and broader patterns of working with agents, including human-in-the-loop review, automated checks and vibe coding. Each term gets a short, plain-English explanation intended to make the underlying concepts learnable quickly.

Website: https://github.com/mattpocock/dictionary-of-ai-coding