Prompt Engineering Glossary
A comprehensive glossary of 30+ terms used in Prompt Engineering.
A
AI (Artificial Intelligence) - Computer systems designed to perform tasks that typically require human intelligence, such as understanding language, recognizing patterns, and making decisions.
AI Assistant - An AI-powered tool designed to help users with tasks through conversation, like ChatGPT, Claude, or Google Gemini.
Audience - The intended readers or users of the content you're creating. Specifying audience helps AI adjust complexity and tone.
B
Brainstorming - Using AI to generate multiple ideas, options, or creative possibilities for a topic or problem.
C
Chain-of-Thought (CoT) - A prompting technique where you ask AI to show its reasoning step-by-step, often improving accuracy for complex problems.
Clarity - Being unambiguous and precise in your prompt so AI doesn't need to guess what you mean.
Claude - An AI assistant developed by Anthropic, known for its helpfulness and focus on safety.
Completion - The response or output that AI generates based on your prompt.
Context - Background information provided in a prompt that helps AI understand your specific situation, needs, and constraints.
Context Window - The maximum amount of text (including your prompt and AI's response) that an AI model can process at once. Larger context windows allow for longer conversations and documents.
D
Default Behavior - What AI does when you don't specify something. Understanding defaults helps you know when you need to be explicit.
E
Editing Prompt - A prompt asking AI to modify existing content—shorten, change tone, fix errors, or improve clarity.
Example (in prompting) - A sample you provide in your prompt to show AI what kind of output you want. See also: Few-Shot Learning.
F
Few-Shot Learning - Providing one or more examples in your prompt so AI can learn the pattern, format, or style you want.
Format - The structure of the output you want—list, table, paragraphs, email, etc. Specifying format gives AI clear structural guidance.
G
Generative AI - AI systems that create new content (text, images, audio, code) rather than just analyzing existing data.
GPT (Generative Pre-trained Transformer) - The architecture behind ChatGPT and similar AI models. Often used informally to refer to AI chatbots.
H
Hallucination - When AI generates false, fabricated, or incorrect information that sounds plausible. Always verify important facts.
I
Input - Your prompt—the text you send to AI. See also: Prompt.
Input-Output Pattern - A prompting technique where you show AI transformation examples (input → output) so it learns to apply the same transformation to new inputs.
Iteration - The process of refining your prompt based on AI's response to get closer to what you actually want. A key skill in prompt engineering.
L
LLM (Large Language Model) - AI models trained on massive amounts of text that can understand and generate human-like text. Examples: GPT-4, Claude, LLaMA.
Length Specification - Defining how long you want the output—in words, sentences, paragraphs, or approximate size.
M
Model - The specific AI system you're interacting with (GPT-4, Claude 3, etc.). Different models have different capabilities and limitations.
N
Natural Language - Ordinary human language (as opposed to programming code). AI assistants are designed to understand and respond in natural language.
O
One-Shot Learning - Providing a single example in your prompt to show AI the pattern you want. See also: Few-Shot Learning.
Output - What AI generates in response to your prompt. See also: Response, Completion.
Overloading - Including too many requirements in a single prompt, which can confuse AI and produce poor results.
P
Persona - A role or character you ask AI to adopt (e.g., "You are an experienced marketing consultant"). Can influence tone and approach.
Prompt - The text input you give to an AI system—your instruction, question, or request that guides what the AI generates.
Prompt Engineering - The skill of crafting effective prompts to get better results from AI systems.
Purpose - Why you need the content and what you want to achieve with it. Clarifying purpose helps AI create more useful output.
R
Refinement - Adjusting your prompt or asking for modifications to improve AI's output. Part of the iteration process.
Response - What AI generates based on your prompt. The quality of the response depends heavily on the quality of your prompt.
Role Prompting - Starting a prompt by defining AI's role ("You are a financial advisor...") to influence the perspective and expertise in the response.
S
Specificity - The level of detail in your prompt. More specific prompts generally produce more relevant and useful results.
System Prompt - Hidden instructions that set the AI's behavior and context. In ChatGPT's custom instructions or when using APIs.
T
Temperature - A setting that controls randomness in AI outputs. Higher temperature = more creative/varied; lower = more focused/deterministic.
Token - The basic unit of text that AI processes. Roughly corresponds to word parts. Important for understanding context window limits.
Tone - The feeling or voice of the content—professional, casual, friendly, serious, etc. Specifying tone helps AI match your communication style.
Transformation Prompt - A prompt asking AI to change content from one form to another—like summarizing, translating, or reformatting.
V
Vague Prompt - A prompt lacking sufficient detail, forcing AI to make assumptions that may not match your intent.
Verification - Checking AI's output for accuracy, especially for facts, data, or important information. Essential because AI can hallucinate.
Z
Zero-Shot - Asking AI to perform a task without providing any examples. The AI relies entirely on its training and your instructions.
Quick Reference by Category
Prompting Techniques
Few-Shot Learning, One-Shot, Zero-Shot, Chain-of-Thought, Role Prompting, Input-Output Pattern
Prompt Elements
Clarity, Context, Specificity, Tone, Format, Length, Audience, Purpose
AI Concepts
LLM, Token, Temperature, Context Window, Hallucination, Completion
Common Issues
Vague Prompt, Overloading, Hallucination
This glossary covers the essential terms for the Introduction to Prompt Engineering course.