What is Prompting?

A prompt is the input we give to an AI agent — it's a combination of instructions and context that helps the AI understand what kind of output we expect. The better and clearer your prompt is, the better and more relevant the response you'll get from the AI.

When working with AI models (like chatbots or generative tools), prompting becomes a powerful skill. There are three main types of prompting:

---

🟢 1. Zero-Shot Prompting

In zero-shot prompting, we ask the AI a question directly without giving it any examples. We're expecting it to understand and generate a response based solely on its existing knowledge.

Example:
"Write a cover letter for a frontend developer job."

---

🟡 2. Few-Shot Prompting

Few-shot prompting means we give the AI a few examples of what we want — usually 1 to 3. These examples help guide the AI toward producing better results.

Example:
You show the AI a design or give it 1–2 samples and say:
"I want something like this, with these features."
You're giving it a reference to follow.

---

🔵 3. Chain-of-Thought Prompting

Chain-of-thought prompting is used when we want the AI to reason step by step. Here, we often give more examples (sometimes 5 to 9 or more), and we also ask the AI to think through the problem logically before answering.

Example:
“Let’s think step by step...”
or
“First we calculate X, then we find Y...”

This is especially helpful for complex reasoning, coding, or multi-step tasks.

👀 Before reading this, did you know there were different types of prompts? Which one was new to you?