Prompt Engineering Introduction

A prompt is what you type into an AI tool. Prompt engineering is the skill of writing prompts that get useful, accurate answers on the first try instead of the fifth. It needs no coding, so it’s useful for everyone: students, developers, teachers, marketers, shop owners.

The same AI can give a vague, generic reply or a precise, ready-to-use one. The difference is almost always the prompt.

The tools#

You’ll use the same skills in all of them:

Tool	Made by	Good to know
ChatGPT	OpenAI	The most widely used chatbot
Claude	Anthropic	Strong at writing, long documents and code
Gemini	Google	Built into Google apps and Android
Microsoft Copilot	Microsoft	Inside Windows, Edge and Office
GitHub Copilot, Cursor	GitHub, Anysphere	AI inside your code editor

Most have a free version that is more than enough to practise this course.

How a chatbot writes an answer#

These tools are built on a large language model (LLM). An LLM was trained on a huge amount of text to do one thing: predict the next piece of text. Pieces of text are called tokens; a token is a word or part of a word. Step through it:

The model reads your prompt, scores every possible next token, picks one, adds it, and repeats until the answer is complete. It doesn’t look facts up in a database (unless the tool adds web search); it continues text in the way that seems most likely given everything it has read.

That explains three things you’ll notice:

1. Your prompt is the starting text. Clear, detailed starting text leads to a better continuation.
2. Answers vary. Ask twice and you may get different wording, because the next token is partly chosen at random. A setting called temperature controls how much.
3. It can sound confident and still be wrong. Producing likely-sounding text isn’t the same as checking facts. This is called hallucination, and you’ll learn to manage it in the last lesson.

Same question, two prompts#

A vague prompt:

Tell me about Python.

You’ll get a general essay that tries to cover everything for nobody in particular.

A better prompt:

I'm a commerce student with no coding experience. In 5 short bullet points,
explain what Python is used for in finance and accounting jobs, with one
real example task for each.

Now the AI knows who you are, what you want, how much and in what shape. The answer is shorter and far more useful.

What this course covers#

1. Writing clear instructions
2. Roles and context
3. Teaching by example (few-shot prompts)
4. Step-by-step thinking and breaking big tasks down
5. Getting tables, lists and JSON back
6. Using AI to learn coding and to work faster
7. Limits, safety and using AI responsibly

Quick check

Why might the same prompt give a different answer each time? Because the model picks each next token with some randomness. It’s normal; ask it to be more specific, or regenerate.

Summary#

• A prompt is your instruction; prompt engineering is writing it well.
• LLMs write by predicting one token at a time, starting from your prompt.
• Clear prompts say who it’s for, what you want, how much and in what format.