Glossary
Plain-English definitions for every key concept in vibe coding, AI-assisted development, and building full-stack software. Each term links to the relevant courses in the curriculum.
An HTTP status code meaning the server cannot find the requested page or resource.
AI systems that autonomously take sequences of actions — reading files, calling tools, making decisions — to complete longer tasks.
Using AI tools to write, review, explain, and debug code — while the developer stays in control of decisions.
Tools that track how users interact with your application, helping you understand behavior, find problems, and measure growth.
Visual transitions and motion effects that make your UI feel responsive, polished, and alive.
A defined interface that lets different software components communicate — the contract between a frontend and backend, or between two services.
The process of verifying who a user is, typically through email/password, social login, or magic links.
The server-side logic of an application — databases, business rules, APIs, and infrastructure.
A task that runs outside the normal request-response cycle, like sending emails, processing uploads, or syncing data.
Storing frequently accessed data in a fast layer (memory, Redis, CDN) to avoid recomputing or refetching it every time.
A prompting technique where you ask the AI to reason step-by-step before giving a final answer, improving accuracy on complex tasks.
Automated pipelines that test your code on every push (Continuous Integration) and deploy it automatically (Continuous Deployment).
A self-contained, reusable piece of UI that manages its own structure, styling, and behaviour.
The maximum amount of text (code, chat history, files) that an AI model can read and use at one time.
An AI-first code editor built on VS Code that embeds language model assistance directly into the editing experience.
An alternative color scheme with light text on dark backgrounds, reducing eye strain and saving battery on OLED screens.
An organised system for storing, querying, and managing structured data that persists beyond a single request.
A data structure that speeds up database queries by letting the database find rows without scanning every record.
The process of identifying and fixing errors, unexpected behaviours, and crashes in your application.
The process of making your application available to real users on the internet.
A collection of reusable UI components, design tokens, and guidelines that keep your application visually consistent.
A platform that packages your application and all its dependencies into a portable container that runs the same everywhere.
Automated emails triggered by user actions, like welcome messages, password resets, and order confirmations.
Numerical representations of text that capture meaning, allowing AI to measure similarity between words, sentences, or documents.
Secret values (API keys, database URLs, credentials) stored outside your code so they never end up in version control.
Cloud services like S3 or Supabase Storage that store and serve user-uploaded files such as images, documents, and media.
Training an existing AI model on your own dataset to specialize its behaviour for a specific task or domain.
The part of a web application that runs in the browser — everything the user sees and interacts with.
Development that covers both the frontend (browser) and backend (server) of an application.
The industry-standard distributed version control system used to track code changes and collaborate on software projects.
When an AI generates confident, plausible-sounding output that is factually wrong or completely made up.
The process of running an AI model to generate output from input. Every AI response is an inference call.
Checking that user-submitted data meets expected rules before processing it, preventing errors and security vulnerabilities.
A test that verifies multiple parts of your application work correctly together.
A lightweight data format used to exchange information between servers, APIs, and front-end applications.
Key combinations that trigger actions in code editors and AI tools, speeding up your development workflow.
A neural network trained on massive text datasets that can understand, generate, and reason about language and code.
Automated code analysis that catches style issues, potential bugs, and inconsistencies before your code runs.
An open standard that lets AI models securely connect to external tools, databases, and services.
A versioned script that changes your database structure, like adding a table or column, in a repeatable way.
Tools and practices for tracking your application's health, performance, and errors in real time.
AI models that can understand and generate multiple types of content: text, images, audio, video, and code.
The simplest version of a product that delivers core value to users, built to validate an idea before investing in full development.
A full-stack React framework that handles routing, server rendering, API routes, and deployment out of the box.
The default package manager for JavaScript and TypeScript, used to install, manage, and share code libraries.
A protocol that lets users sign in to your app using their existing Google, GitHub, or other third-party accounts.
A tool that lets you interact with a database using your programming language instead of writing raw SQL queries.
Connecting your app to a payment processor like Stripe or Polar to accept subscriptions, one-time payments, or invoices.
An automatic temporary deployment of a pull request branch, letting you test and share changes before merging.
The skill of writing effective instructions that get AI models to produce the output you actually want.
A request for specific data from a database, API, or search engine, usually written in a structured language like SQL or GraphQL.
A technique where an AI retrieves relevant information from an external knowledge base before generating its answer.
Restricting how many requests a user or IP address can make in a given time period to prevent abuse and protect your API.
A JavaScript library for building user interfaces from reusable components, maintained by Meta and used by millions of developers.
Techniques for pushing live updates from the server to the browser without the user refreshing the page.
Restructuring existing code to improve clarity, maintainability, or performance without changing its external behaviour.
Building interfaces that adapt gracefully to any screen size, from mobile phones to wide desktop monitors.
Reverting a deployment or database change to a previous working version when something goes wrong.
Software as a Service — a business model where users pay a recurring subscription to access software hosted online.
Generating the initial structure and boilerplate of a project, feature, or component so you can start building on a working foundation.
A systematic review of your application's code, configuration, and dependencies to find and fix vulnerabilities.
Pre-populated test records inserted into a database during development to simulate real-world usage.
Generating HTML on the server for each request, so the page arrives ready to display instead of loading as a blank shell.
A deployment model where your code runs on demand without managing servers. You pay per execution, not per hour.
Displaying AI-generated text token-by-token as it is produced, rather than waiting for the entire response to complete.
Hidden instructions given to an AI model that define its behaviour, tone, and constraints before the user's first message.
A utility-first CSS framework where you style elements by applying small, composable classes directly in HTML.
Shortcuts and compromises in code that ship faster now but create extra work later when the codebase needs to change.
A setting that controls how creative or deterministic an AI model's output is. Lower values give predictable answers; higher values increase randomness.
The smallest unit of text an LLM processes. Roughly one token per word, but punctuation, code symbols, and subwords each count separately.
The ability of an AI model to call external functions, APIs, or tools as part of generating a response.
A typed superset of JavaScript that catches errors at development time, making code safer and easier for both humans and AI to work with.
The visual design (UI) and overall feel (UX) of an application, covering everything from button placement to user flow.
A small, isolated test that verifies a single function or module behaves correctly.
A system that tracks every change to your codebase, so you can review history, collaborate, and undo mistakes.
Building software by directing AI through natural language instead of writing every line by hand.
An HTTP callback that lets external services notify your app when something happens, like a payment completing or a form submission.
A structured text format for encoding documents and data, still used in configuration files, RSS feeds, and enterprise systems.
A human-readable configuration format commonly used for CI/CD pipelines, Docker Compose files, and infrastructure-as-code.
Prompting techniques where you give the AI zero or a few examples of the desired output before asking it to perform a task.