In partnership with

You’re brilliant on calls. Why do your emails sound flat?

I built a system that turns one coaching call into weeks of emails using your actual words. No blank page. No AI slop. Just your real thinking, extracted.

Turn AI into Your Income Engine

Ready to transform artificial intelligence from a buzzword into your personal revenue generator?

HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.

Inside you'll discover:

  • A curated collection of 200+ profitable opportunities spanning content creation, e-commerce, gaming, and emerging digital markets—each vetted for real-world potential

  • Step-by-step implementation guides designed for beginners, making AI accessible regardless of your technical background

  • Cutting-edge strategies aligned with current market trends, ensuring your ventures stay ahead of the curve

Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.

This week I was building a big presentation for a client.

And somewhere in the middle of it, we made a deliberate decision: for the more complex pieces, we'd stop trying to write the perfect prompt ourselves.

Instead, we'd tell AI what output we needed and ask it to write the prompt for us.

Not for everything. For the simple stuff, you can just ask and go.

But for the bigger deliverables where quality and consistency matter? Letting AI build the prompt makes it way easier to get to the outcome you want.

Here's why: AI knows what it needs better than you do.

When you write a prompt yourself, you're guessing at what details to include, what format to specify, what context matters. You're reverse engineering a language you didn't build.

But when you say something like "I need a 10-slide consulting deck that does A, B, and C for this specific persona. Create a prompt to get the info you need to build a detailed outline with everything necessary," the AI fills in gaps you didn't even know existed. Claude, ChatGPT, Gemini, doesn't matter which one. They all do this well.

Result?

Better output. Faster. And repeatable.

How to do it

Here's the simple version:

1️⃣ Describe the output you want. Be specific about what "done" looks like. Show an example if you have one.

2️⃣ Ask AI to create a prompt that will produce that output consistently. Say exactly that: "Create a prompt I can reuse to get this result every time."

3️⃣ Run the prompt. See what you get.

4️⃣ Edit the output. Make it yours. Fix what's off, add what's missing.

5️⃣ Give the edited version back to AI and say: "Now reverse engineer this final version and update the prompt so it produces this output."

That last step is where the magic is. You're training the prompt with your actual standards, not your best guess at describing them.

And if you a better prompt to make prompts, here's a prompt you can use to have AI write prompts for you:

You are an expert prompt engineer. Your job is to help me create a highly effective AI prompt for a specific scenario.

Here's how this works:

  1. I'll describe what I'm trying to accomplish.

  2. You'll ask me up to 5 clarifying questions to fill in gaps (audience, tone, format, constraints, desired output).

  3. Once you have enough context, generate a complete, ready-to-use prompt that includes:

  • A clear role assignment for the AI

  • Specific context about the task

  • Explicit instructions on format, length, and tone

  • Any constraints or guardrails

  • A concrete example of what good output looks like (when helpful)

  • A quality checklist the AI should self-check against before delivering

Write the prompt in plain language. No jargon. Structure it so the AI knows exactly what "done" looks like.

After delivering the prompt, suggest 2-3 ways I could tweak it for better results depending on what I'm optimizing for (speed, depth, creativity, persuasion, etc.).

My scenario: [DESCRIBE WHAT YOU WANT HERE]

Confession: I got this by asking AI "help me create a prompt that would get AI to create a great prompt for a specific scenario." 😀

My Take

Remember when "prompt engineer" was a job title? Companies were hiring people at six figures to figure out how to talk to AI.

I'm curious what happened to those roles. Because the skill isn't really about crafting the perfect prompt anymore. It's about knowing what output you need and letting AI do the prompt engineering for you.

The other thing to keep in mind: these prompts become intellectual property.

Think about it. When you go through the pain of describing your ideal output, running it, editing it, feeding it back, and refining the prompt until it nails your standard every time? That prompt is now proprietary.

It's not just a string of text. It's a codified system for producing high-quality work consistently and ithout starting from scratch.

For this presentation I’m working on, we're building a proprietary prompt library.

A collection of prompts that produce specific deliverables, in our voice, at our quality bar. Every time.

That's not a shortcut. That's an asset.

So if you're still sitting there agonizing over how to word your prompts, try flipping it.

Tell AI what you want to end up with. Let it write the prompt. Then refine from there.

Dig Deeper

👉 Check out this issue where I walked through creating a prompt to extract a framework from scratch. First step? Having AI create a prompt: I Used AI to Analyze How I Use AI

What did you think of today's edition?

Login or Subscribe to participate

Thanks for reading!

Nathan Rodgers

👋 Say hello on Substack and LinkedIn

Were you forwarded this email? Click here to subscribe.

Keep Reading