AI Glossary

Agent Autonomy

How much freedom an AI agent has over its own next action, ranging from fully scripted to fully autonomous, not a simple on/off switch.

AI Agent

An AI system that perceives its environment, makes decisions, and takes actions autonomously to achieve a goal.

API Key

A secret string of characters that identifies your account when your code calls an external service like an AI model.

Chain of Thought

A prompting technique where a model reasons step by step before giving its final answer or next action, instead of jumping straight to a conclusion.

Chatbot

A software program that simulates conversation: from simple rule-based bots to LLM-powered assistants like ChatGPT.

Claude Code

Anthropic's official command-line tool that lets you use Claude to write, edit, and reason about code directly in your terminal.

Command Line

The text prompt inside a terminal where you type commands: ls, cd, git, npm, to control your computer.

Context Window

The maximum amount of text an LLM can read and reason over at once, measured in tokens.

Cursor

An AI-powered code editor built on VS Code that has Chat, Inline Edit, and Agent modes for writing and editing code with AI assistance.

Embedding

A list of numbers that captures the meaning of a word, sentence, or document so that similar meanings land close together in space.

Environment Variable

A named value stored outside your code, like an API key or database URL, that your program reads at runtime.

Fine-tuning

Additional training of a pre-trained model on a smaller, task-specific dataset to improve its performance on that task.

Generative AI

AI systems that generate new content, text, images, code, audio, rather than just classifying or predicting from existing data.

Git

The most widely used version control system: a tool that tracks changes to your files locally on your computer.

GitHub

A website where developers store, share, and collaborate on code that is tracked by Git.

Hallucination

When an AI model generates text that sounds confident and plausible but is factually wrong or completely made up.

Inference

Running a trained AI model to generate an output: what happens every time you send a message to an AI.

Large Language Model (LLM)

A neural network trained on vast amounts of text that can understand and generate human language.

Machine Learning

A branch of AI where systems learn patterns from data rather than being explicitly programmed with rules.

MCP (Model Context Protocol)

An open standard that lets AI models connect to tools, data sources, and apps through one shared protocol instead of a custom integration for every model-tool pairing.

Model

In AI, a model is a trained system that takes an input and produces an output: the core artifact produced by machine learning.

Multi-Agent System

An architecture where multiple AI agents work together, each handling a specialized task, to complete goals too complex for a single agent.

Neural Network

A computational system loosely inspired by the brain: layers of interconnected nodes that learn patterns from data.

Node.js

A runtime that lets JavaScript run outside the browser, on your computer or a server, making it possible to build backend tools and AI apps with JavaScript.

Orchestration (Multi-Agent Orchestration)

Coordinating two or more AI agents, each with a distinct role and prompt, so they complete together a task one agent would handle worse alone.

Package Manager

A tool that downloads, installs, and manages third-party code libraries so you don't have to do it manually.

Parameter

A number inside a neural network that is adjusted during training: '7 billion parameters' means the model has 7 billion such numbers.

PATH (Environment Variable)

A list of directories your computer searches when you type a command, if Python or Git says 'command not found', PATH is usually why.

Prompt

The input text you give to an AI model: instructions, context, examples, and questions combined.

Prompt Engineering

The practice of crafting inputs to AI models to get better, more accurate, or more useful outputs.

Python

A beginner-friendly programming language that dominates AI and data science because of its readable syntax and vast library ecosystem.

RAG (Retrieval-Augmented Generation)

A technique that gives an LLM access to external documents at query time so its answers are grounded in up-to-date or private information.

ReAct (Reason + Act)

An agent pattern where the model alternates between reasoning and a tool call, reading back the result before deciding what to do next.

System Prompt

Instructions given to an AI model before the conversation starts that define its role, tone, and constraints.

Temperature

A setting that controls how random or creative an AI model's outputs are: low temperature is more predictable, high is more creative.

Terminal

A text-based interface where you type commands to control your computer directly, without clicking buttons or menus.

Token

The basic unit an LLM reads and writes, roughly a word or part of a word, depending on the tokenizer.

Tool Use

The ability of an AI model to call external functions, like searching the web, running code, or querying a database, during a conversation.

Transformer

The neural network architecture introduced in 2017 that powers virtually all modern large language models, including GPT, Claude, and Gemini.

Vector Store (Vector Database)

A database built to store embeddings and quickly find the ones closest in meaning to a query, powering retrieval in RAG systems.

Version Control

A system that tracks every change made to files over time, letting you revert mistakes and collaborate without overwriting each other's work.