Prompt

What Is a Prompt?

A prompt is everything you send to an AI model before it generates a response. It can be as simple as a single question or as complex as a multi-thousand-word document containing a role description, detailed instructions, background context, worked examples, and the actual task.

The model sees nothing except what is in the prompt — which is why what you put in, and how you structure it, matters enormously.

Anatomy of a Prompt

Most API calls to an LLM have three parts:

• System message — sets the model's role, persona, and ground rules ("You are a helpful customer support agent for Acme Corp. Be concise and friendly.")
• User message — the actual request or question
• Examples (few-shot) — optional worked examples that show the model the pattern you want it to follow

Together these form a structured input that shapes the model's entire response.

Why Prompts Matter So Much

LLMs are sensitive to phrasing. Asking "Summarize this" versus "Summarize this in three bullet points for a non-technical executive" can produce dramatically different outputs. A vague prompt invites vague answers; a specific prompt tends to produce specific, useful ones.

Prompt Engineering as a Skill

Prompt engineering is the practice of designing, testing, and iterating on prompts to get reliable, high-quality outputs from AI models. It includes techniques like:

• Chain-of-thought — asking the model to reason step by step before answering
• Few-shot prompting — providing examples of good inputs and outputs
• Role assignment — giving the model a persona to adopt

It is a learnable skill that pays off quickly, even for non-developers.