Dictionary

Media queries are a CSS feature that allows you to apply styles conditionally based on device characteristics like viewport width, height, orientation, color scheme preference, or reduced motion settings. They are the backbone of responsive web design, enabling different layouts for mobile, tablet, and desktop screens. Modern media queries also support user preference detection with `prefers-color-scheme` and `prefers-reduced-motion` for more accessible experiences.

Memoization in React is a performance optimization that caches the result of expensive computations or component renders to avoid redundant work on re-renders. React provides `React.memo()` to skip re-rendering a component when its props haven't changed, `useMemo()` to cache computed values, and `useCallback()` to cache function references. While powerful, premature memoization adds complexity — it's best applied after profiling confirms an actual performance bottleneck.

A microinteraction is a small, contained moment in a user interface that accomplishes a single task — a button animation on click, a toggle switch, a pull-to-refresh gesture, or a "like" heart animation. These subtle design details provide feedback, guide users, and add personality to an interface. Well-crafted microinteractions make products feel polished and responsive without overwhelming users.

A modal (or modal dialog) is a UI element that overlays the main content to demand user attention and interaction before they can return to the underlying page. Modals are used for confirmations, forms, alerts, and content previews. Proper implementation requires focus trapping, keyboard accessibility (Escape to close), and preventing background scroll — making accessible modals more complex than they appear.

Model Context Protocol (MCP) is an open standard developed by Anthropic that defines how AI applications connect to external data sources and tools. MCP provides a universal interface for LLMs to access databases, APIs, file systems, and other services through standardized server implementations. It enables building AI applications that can interact with the real world in a structured, secure way.

Multimodal AI refers to models that can process and generate multiple types of data — such as text, images, audio, and video — within a single system. Models like GPT-4o and Claude can accept both text and image inputs, enabling use cases like visual question answering, document analysis, and UI understanding. This convergence is blurring the lines between previously separate AI disciplines.