Glossary

Every technical term from the course, defined in plain English. Organized by the lesson where each concept is introduced.


Lesson 1: Your Computer Is Two Things
Compute
The processing power of a computer: the CPU running instructions, performing calculations, executing functions.
Storage
Persistent data that survives when the power goes off. Hard drives, SSDs, databases.
RAM
Random Access Memory. Fast, temporary storage that gets erased when the power goes off. Like the bed of a pickup truck: small, fast, holds what you are actively working on.
Operating System (OS)
The software layer (Windows, macOS, Linux) that provides a user-friendly interface for managing compute, storage, and networking.
Networking
The ability for computers to communicate with each other over a network (WiFi, Ethernet, the internet).
Lesson 2: The Terminal & Claude Code
Terminal
A text-based interface for interacting with your computer by typing commands instead of clicking. The same thing as Finder, just with text.
IDE
Integrated Development Environment. A program combining a file browser, code editor, and terminal in one window (e.g., VS Code, Cursor).
File Path
The address of a file on your computer, e.g., /Users/you/Documents/my-project/index.html. Each slash is a folder boundary.
Claude Code
An AI agent that runs inside your terminal. It reads files, writes code, runs commands, and deploys changes directly in your project.
Lesson 3: Your First Vibe-Code
Vibe Coding
Describing what you want in plain English and having AI generate the code. The workflow of describe, watch, adjust, repeat.
Debugging Loop
The workflow of: something breaks, copy the error, paste it to Claude, watch it fix. Repeat until it passes.
Lesson 4: CLAUDE.md: Session Memory
CLAUDE.md
A plain text file in the root of your project that Claude Code reads at the start of every session. Persistent project memory, like the glossary in front of a textbook.
/init
A Claude Code slash command that scans your project and generates a starter CLAUDE.md automatically.
Lesson 5: CRUD Is Everything
CRUD
Create, Read, Update, Delete. The only four operations you can ever do with data. No exceptions.
Function
A set of instructions that performs one or more CRUD operations. The 'how' of computing.
Trigger
What kicks off a function: manual (click), cron (timer), webhook (external message), or event-driven (data change).
Cron
A scheduled trigger that fires at set times. From the Greek word for time (chronos).
Webhook
An external service sending your system a notification that something happened.
Lesson 6: Files & The Browser
HTML
HyperText Markup Language. Text with tags that browsers render into visual web pages.
CSS
Cascading Style Sheets. A text file that tells the browser how HTML elements should look (colors, fonts, spacing).
Markdown
A lightweight text format using symbols (# for headings, ** for bold) that renderers turn into formatted text.
JSON
JavaScript Object Notation. A structured data format with key-value pairs. The lingua franca of web APIs. Readable by a baby.
Rendering
The process of interpreting raw data (like HTML tags) and displaying it visually. Chrome is a renderer.
Lesson 7: APIs: The Restaurant Analogy
API
Application Programming Interface. A contract between two systems: one asks, the other answers. Like a restaurant menu.
Base URL
The street address of the restaurant, meaning the root address of an API server.
Endpoint
The menu item. A specific path on an API that performs a specific action.
Parameters
Your order details. The specific data you send with an API request (coordinates, date range, filters).
Request
A message sent to a server asking for data or an action.
Response
What the server sends back after processing a request.
HTTP Method
The verb in an API request: GET (read), POST (create), PUT (update), DELETE (delete).
Status Code
A three-digit number indicating what happened. 200 good, 404 bad, 500 server broke.
API Key
An unauthenticated bearer token, a secret string included with requests. Whoever has the key can use it.
.env File
An environment file that stores secrets (API keys, passwords) outside your code. Never committed to version control.
.gitignore
A file that tells Git which files to never track or upload. Used to keep secrets out of repositories.
Rate Limiting
APIs throttling how many requests you can make in a time period to prevent abuse.
Lesson 8: MCP: Teaching Claude to Use APIs
MCP
Model Context Protocol. A standardized way to give AI models pre-built instruction manuals for APIs and tools. An API is a door; MCP is the instruction manual for that door.
MCP Server
A service that implements MCP for a specific tool or API, giving Claude named tools with defined parameters instead of making it guess.
Lesson 9: The Harness
Harness
The connection layer that wires Claude to your project, tools, MCPs, skills, and codebase. The model reasons; the harness acts.
Agentic Loop
The cycle Claude Code runs: gather context, take action, verify results, repeat. Powered by the harness.
Slash Command
A command prefixed with / that talks to the harness, not the model. Examples: /init, /mcp, /clear, /resume, /usage, /remote.
/mcp
Shows which MCP servers are connected and what tools they provide in the current session.
/usage
Shows token usage and billing status for the current Claude Code session.
/remote
Runs a Claude Code session on a remote server instead of your local machine.
Lesson 10: Skills
Skill
A reusable instruction set (text file) that teaches Claude a specific workflow. MCP teaches how to use a tool; a skill teaches how to do a job.
Lesson 11: Scheduled Agents
Scheduled Agent
Claude Code running on a timer, automatically, in the background, on a recurring schedule. Like a cron job but with AI.
Lesson 12: Code: What It Actually Is
Variable
A labeled container that holds a piece of data: a name, a number, true/false.
Loop
Code that repeats the same instruction multiple times, once for each item in a list.
If/Else
How computers make decisions. Check a condition: if true, do one thing; if false, do another.
Codebase
All the code files for a project, organized into folders.
TypeScript
JavaScript with type-checking guardrails that catch mistakes earlier.
Python
A programming language popular for data, AI, and scripting. Reads almost like English.
Lesson 13: How Code Actually Runs
Compiled Language
A language (Go, Rust, C) translated entirely to machine code before running. Like translating a whole book before publishing.
Interpreted Language
A language (Python, JavaScript) translated line-by-line at runtime. Like a live interpreter translating as someone speaks.
Runtime
The environment that makes code run: the interpreter plus memory management, file access, etc.
Static Site
Pre-built HTML files served as-is. No code runs at request time.
SSR
Server-Side Rendering. Code runs on the server for every request to build fresh HTML.
SSG
Static Site Generation. Code runs once at deploy time to generate all HTML pages.
Lesson 14: Repos & Version Control
Repository (Repo)
A folder of code with full version history. Every change ever made is recorded.
Git
The version control tool that tracks changes to code. A time machine for your project.
Commit
A snapshot, a save point in your code's history, with a message explaining what changed.
Branch
A parallel copy of the codebase where you can work on changes without affecting the main version.
Pull Request (PR)
A request to merge your branch's changes into the main branch, with review and discussion.
CI/CD
Continuous Integration / Continuous Deployment. Automated pipelines that test and deploy code when it's merged.
GitHub
A website that hosts Git repositories online, with collaboration features like PRs, issues, and Actions.
Lesson 15: Databases & Storage
Database
An organized collection of data with rules, like a spreadsheet with enforced structure.
SQL
Structured Query Language. The language for talking to relational databases. Reads like English.
Table
A single collection of related data in a database, like one tab in a spreadsheet.
Relational Database
A database (like PostgreSQL) with strict tables where data references other tables.
Document Database
A database (like MongoDB, Firestore) that stores flexible JSON-like documents.
Cache
Frequently-used data kept in fast memory so you don't have to fetch it from the database every time.
Lesson 16: Frameworks & Libraries
Framework
A pre-built structure that handles common patterns so you focus on your specific logic.
Library
A collection of reusable code that does one thing well (e.g., date formatting, chart rendering).
Stack
The combination of technologies used to build an app: language, framework, database, hosting.
package.json
A file listing a JavaScript project's dependencies and scripts. The recipe card.
Next.js
A React framework for building web apps with server-side rendering and routing.
Lesson 17: The Cloud: Hardware Stores for Computers
Cloud
Other people's computers in a warehouse, sliced up so you rent exactly the pieces you need. Hardware stores for computers.
Bare Metal
A physical server you rent entirely. Fixed cost whether you use 1% or 100%.
AWS
Amazon Web Services. The largest cloud provider.
GCP
Google Cloud Platform. Strong in AI/ML and developer tooling.
Azure
Microsoft's cloud. Tightly integrated with the Microsoft ecosystem.
Lesson 18: Hosting & Serving
Server
A computer running code, waiting for requests. Your laptop could be one.
Serverless
Cloud hosting where you don't manage servers. Code runs only when needed, and you pay per execution. Uber vs. owning a car.
Container
Everything your code needs to run, packed into one portable box (runtime, libraries, config, code).
Docker
The most common tool for building containers. A Dockerfile is the recipe.
CDN
Content Delivery Network. Servers distributed globally that serve files from the location nearest to the user.
Cold Start
The delay when a serverless container needs to wake up after being idle.
Lesson 19: AI & Language Models
LLM
Large Language Model. A program that predicts the next word to generate text responses. Autocomplete on steroids.
Token
The unit an LLM processes, roughly 3/4 of a word. Both input and output are measured in tokens.
Temperature
Controls randomness in AI output. 0 = deterministic and factual, 1 = creative and varied.
Context Window
The maximum amount of text an LLM can consider at once, like the size of its working desk.
Prompt
The instruction you give an LLM. The input text that shapes the output.