A low-fidelity visual representation of a page layout focusing on structure and content hierarchy without detailed styling. Wireframes help validate concepts quickly before investing in visual design. In Framer, you can start with simple shapes and progressively add detail as designs evolve.

Related AI terms: First Draft and Prompt-to-Code.

A Sandbox is a constrained runtime environment that limits file, network, or system access to reduce risk during automated execution.

Token Budget is the practical limit you set for prompt and completion tokens to balance quality, latency, and cost.

A Prompt Template is a standardized prompt format that injects variable content into fixed instructions for consistent outputs.

Related AI terms: Prompt Enhancement and Prompt-to-Code.

First Draft is an AI workflow for quickly generating editable interface concepts. Teams use it to move quickly from idea to wireframe and then refine output with Prompt-to-Code.

Prompt Enhancement is an AI-assisted rewrite step that expands or clarifies user intent before generation. It often works with a Prompt Template and explicit System Prompt constraints.

A development approach coined by Andrej Karpathy in which developers describe what they want in plain language—or even just a vibe—and rely on AI to write, iterate, and debug the code. Rather than authoring every line, the developer acts as director and reviewer, accepting or rejecting AI suggestions. Vibe coding lowers the barrier to building software and accelerates prototyping, but it requires careful review because AI-generated code can introduce subtle bugs or security issues.

The skill of writing, structuring, and iterating on instructions—often called prompts—so that an AI model produces the desired output. In coding contexts, good prompt engineering includes providing context, specifying constraints, and showing examples. In vibe coding workflows, clear prompts directly determine code quality. Effective prompt engineering reduces hallucinations, improves specificity, and is essential for getting reliable results from both code generation and agentic tasks.

The ability of AI models to translate intent—expressed as prose, pseudocode, or examples—into executable source code. Code generation underpins vibe coding, autocomplete tools, and agentic workflows. Generated code should always be reviewed for correctness, security, and maintainability, since models can produce plausible-looking but incorrect implementations.

Writing software by expressing intent in conversational language rather than formal syntax. AI models translate the natural-language description into runnable code. Natural language programming is the conceptual foundation of vibe coding and is enabled by advances in large language models. It dramatically lowers the barrier to entry for non-developers, though it still requires understanding of what good code looks like in order to review AI output effectively.