Dictionary of AI Coding

A plain-English glossary of AI coding terminology by Matt Pocock: over 100 terms from tokens and context windows to failure modes and handoffs.

Dictionary of AI Coding

Overview

Each entry defines a term in plain language with usage examples and related terms. It is organised in seven sections: the model (tokens, parameters, inference), sessions, context windows and turns, tools and environment, failure modes, handoffs, memory and steering, and patterns of work.

Useful as a shared vocabulary for working with coding agents. Notable entries include "attention degradation" (why agents get worse as a session grows) and "prefix cache" (why repeated context is cheaper).