Prompt Engineering Basics With Examples
Learn how to write clear instructions that get better answers from AI
Key Takeaways
- A prompt is the instruction you give to an AI. Better prompts give better answers.
- A good prompt says who the AI should act as, what to do, the background, the format, and the limits.
- Examples inside your prompt are one of the easiest ways to improve results.
- Always review the answer. You can improve it by asking follow-up questions.
What Is Prompt Engineering?
Prompt engineering means writing instructions so that an AI model understands exactly what you want. It is like giving directions to a new team member. If you say “make a report,” you will get a guess. If you say who the report is for, what it should include, and how long it should be, you get something much more useful.
You do not need to be a programmer to write good prompts. But developers who build AI apps also use prompt engineering every day, because the prompt controls how the app behaves.
Why Prompts Matter
AI models cannot read your mind. They only see the words you give them. A short, unclear prompt often leads to a general or off-topic answer. A clear prompt saves time because you need fewer corrections.
A Simple 5-Part Prompt Formula
When you write a prompt, think about these five parts.
- Role: Who should the AI act as? Example: “You are a friendly maths tutor.”
- Task: What exactly should it do? Example: “Explain fractions to a 12-year-old.”
- Context: What background does it need? Example: “The student finds fractions confusing.”
- Format: How should the answer look? Example: “Use three short steps and one example.”
- Limits: What should it avoid or follow? Example: “Use simple words and keep it under 150 words.”
You do not need all five every time. For small tasks, one or two parts may be enough. For important tasks, use all five.
Weak Prompt vs Strong Prompt
See the difference in these examples.
Example 1: Study help
- Weak: Explain machine learning.
- Strong: Explain machine learning to a first-year BCA student in simple English. Use one real-life example and finish with three key points.
Example 2: Marketing
- Weak: Write a post about our course.
- Strong: Write a LinkedIn post of about 100 words announcing a 3-month Python course for beginners. Use a friendly tone and end with a question to start a discussion.
Example 3: Coding
- Weak: Fix my code.
- Strong: This Python function should return the average of a list, but it fails when the list is empty. Explain the cause and show a corrected version with a short comment.
The strong versions name the audience, the goal, the format, and the length. That is why they work better.
Useful Prompt Techniques
- Be specific: Say exactly what you want, who it is for, and how long it should be.
- Give examples: Show one or two samples of the style or format you like. This is often called few-shot prompting.
- Ask for steps: For problems and math, ask the AI to explain its steps so you can check them.
- Set the format: Ask for a table, a list, a short email, or structured output such as JSON.
- Break big tasks into parts: Ask for an outline first, then ask for each section.
- Improve step by step: If the first answer is not right, say what to change: “Make it shorter and more formal.”
- Let the AI ask questions: Try: “Ask me up to three questions before you write.”
Common Mistakes to Avoid
- Writing prompts that are too short or vague.
- Asking for many different things in one message.
- Not stating the audience or the purpose.
- Trusting the first answer without checking facts.
- Pasting private or sensitive data into public AI tools.
Prompt Engineering for Developers
If you build apps with AI, prompts become part of your software. Developers often work with:
- System prompts: Fixed instructions that set the AI’s role and rules for every user.
- Structured outputs: Asking the model to reply in a set format, such as JSON, so a program can use the result.
- Conversation history: Sending earlier messages so the AI keeps context.
- Tool calling: Letting the AI request actions, like looking up a record.
- Testing: Trying many sample questions to see where the prompt fails.
- Safety: Protecting the app from prompt injection, where a user or document tries to override your instructions.
A prompt is never finished after one try. Teams test it, change it, and test it again.
Try This Practice Exercise
- Pick a simple task, such as writing a short email asking for a leave day.
- Write a weak prompt and save the answer.
- Rewrite the prompt using the five-part formula.
- Compare the two answers.
- Change one part of the prompt, such as the format, and compare again.
Doing this for a week will teach you more than reading ten articles.
Frequently Asked Questions
Prompt writing is now a useful skill in many jobs. By itself it is rarely enough for a technical role, so combine it with Python, APIs, and project work.
Not for everyday use. To build AI applications, you will need coding skills.
Python is very helpful
Not always. Different models respond differently, so test your prompt in the tool you use.
, because most RAG tools and tutorials use it.
As long as it needs to be clear. Short for simple tasks, longer for complex ones.
Learn Prompting and AI Development in Mohali
Prompt engineering is the first step toward building real AI applications. If you are looking for AI training in Mohali, WebX Technology teaches prompting, Python and LLM APIs as part of its generative AI learning path. Visit webxtechnology.org to see the syllabus and choose a duration that suits you.
