Get startedGet started for free

Prompt Engineering with AI

1. Prompt Engineering with AI

Let's now talk about prompt engineering—a term that sounds technical but really comes down to something you already do every day: clear communication.

2. What Prompt Engineering Really Is

Let’s start by busting a few myths. Prompt engineering isn’t coding,

3. What Prompt Engineering Really Is

and it doesn’t require technical skills to do.

4. What Prompt Engineering Really Is

It’s not about magic words or secret formulas. At its core, it’s about three principles of good communication:

5. What Prompt Engineering Really Is

clarity, completeness, and context.

6. What Prompt Engineering Really Is

Imagine you’re explaining a task to a new colleague. They’re smart and capable but don’t share your background or company knowledge. To get great results, you’d need to be clear about what you want, complete in your instructions, and provide context for why it matters. That’s exactly how to think about communicating with AI. This approach helps overcome the curse of knowledge—the assumption that what’s obvious to you is obvious to everyone else.

7. The Four-Part Framework

Now let me introduce a simple four-part framework that will transform how you write prompts: Ask, Requirements, Context, and Examples. Let's break each one down with a real scenario.

8. The Four-Part Framework

Imagine you need to summarize a quarterly report for your executive team. First,

9. The Four-Part Framework

the Ask—this is your core request, stated clearly. For example: 'Summarize this quarterly report for executives.' Simple and direct. Second,

10. The Four-Part Framework

Requirements—these are your specific constraints and preferences. You might say: 'Keep it to two paragraphs, focus on actionable insights rather than raw data, and avoid technical jargon.' Notice how this gives AI clear boundaries to work within. Third,

11. The Four-Part Framework

Context—this is the background information that shapes how the task should be done. 'The executives reading this are non-technical and care most about strategic implications for decision-making.' This helps AI understand the audience and adjust its approach. Finally,

12. The Four-Part Framework

Examples—showing what "good" looks like. You might provide a snippet of a previous summary that hit the mark, or even show a bad example with notes on what to avoid. Examples are incredibly powerful because they demonstrate your expectations in a way that descriptions alone cannot.

13. Applying the Framework

Let’s see the difference this makes. Suppose someone writes a one-line prompt: “Make this better.” What’s missing?

14. Applying the Framework

Everything. There’s no clear goal, no requirements, no context, no example—so the AI has to guess.

15. Applying the Framework

Now apply the framework, starting with the ask: rewrite this product description to be more engaging.

16. Applying the Framework

Requirements: Keep it to a single paragraph, use active voice, and highlight customer benefits over features.

17. Applying the Framework

Context: This is for an e-commerce site where visitors spend an average of eight seconds scanning descriptions.

18. Applying the Framework

And example: Here’s a version that performed well. The second prompt gives the AI exactly what it needs to succeed: purpose, structure, and guidance.

19. The Key Question

Here’s a simple test for any prompt you write: ask yourself, “If I gave these instructions to a smart junior teammate, could they complete the task successfully?” If not—if they’d come back with questions—your prompt needs more work. This exercise builds empathy into your prompt writing. It helps you step outside your own expertise and identify what information or examples are missing. Remember, AI is capable but context-blind. It doesn’t know your organization, your audience, or your unspoken preferences. When your prompts are clear enough that a new teammate could complete the task, AI can, too—often faster and more consistently. The key is giving it the same complete, thoughtful guidance you’d give a human collaborator.

20. Let's practice!

Time to get prompting!

Create Your Free Account

or

By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.