Prompting vocabulary
Prompt clearly. 16 hand-written terms with real workplace examples.
Learn prompting terms every day.
One hand-written term. A real workplace example. Thirty seconds before your day gets loud.
System Prompt
A set of instructions given to an AI before the conversation starts. It tells the AI how to behave for the entire session.
Few-Shot Example
One or more examples inside a prompt that show the AI exactly what output format you want. More examples means more reliable results.
Zero-Shot
A prompt with no examples at all. You give instructions and let the AI figure out the format on its own.
Chain of Thought
Asking the AI to reason step by step before giving its final answer. It reduces mistakes on tasks that involve multiple steps or logic.
Persona
Telling the AI to act as a specific role or expert. The persona shapes how it explains things and what assumptions it makes.
Constraint
A hard rule you add to a prompt that limits what the AI is allowed to do. Constraints prevent the AI from going off-script.
Output Format
Telling the AI exactly how to structure its response. You can request JSON, a list, a table, code only, or any other format.
Iteration
Sending a follow-up prompt to refine a previous result. Each round gets you closer to exactly what you want.
Reference File
A file, URL, or code block you paste into a prompt so the AI has real context to work from instead of guessing.
Hallucination
When an AI states something confidently that is not true. It can invent function names, package names, or facts that do not exist.
Context Window
The maximum amount of text an AI model can read at one time. Files, chat history, and instructions all count toward this limit.
Guardrail
A rule that prevents an AI from producing harmful or off-topic output. Guardrails can be built into the model or added via your system prompt.
Eval
A scored test suite for AI outputs. You run prompts against real examples and track accuracy before shipping changes.
RAG
Retrieval-Augmented Generation. Fetch trusted documents first, then generate an answer grounded in those sources.
Vector Database
A database optimized to store embeddings and find the closest matches by meaning, not exact keywords.
Embedding
A numeric vector that captures the meaning of text. Similar phrases get similar vectors so search can match by concept.
Building toward a role? Browse career pathways that map terms to jobs.