Guide 08

Prompt Engineering Basics

Master the art of asking AI the right questions to get the best answers.

← Back to All Guides

The Basics

What is prompt engineering?

"Prompt engineering" sounds technical, but it really just means learning how to talk to AI so it gives you better answers. Think of it like learning to give good instructions. When you tell someone "make me food," you'll get something random. When you say "make me a grilled cheese sandwich with sourdough bread and cheddar," you get exactly what you want.

The same principle applies to AI. Better instructions = better results. This guide teaches you the fundamentals.

Framework

The 5 building blocks of a great prompt

You don't need all five every time, but the more you include, the better your results will be.

1. Role — Tell AI who to be

"You are a friendly fitness coach" or "Act as an experienced editor." Giving AI a role changes how it responds—its tone, vocabulary, and focus all shift.

Example: "You are a patient math tutor for a 10-year-old."

2. Context — Explain the situation

Give background information. What's the project? Who's the audience? What have you already tried? Context helps AI understand your actual needs instead of guessing.

Example: "I run a small bakery and need to write a social media post about our new gluten-free menu."

3. Task — Say what you want

Be crystal clear about the action. "Write," "List," "Summarize," "Compare," "Explain." Use specific action words so there's no ambiguity about what you need.

Example: "Write 3 variations of an Instagram caption for this post."

4. Format — Describe the output

Tell AI exactly how you want the response: a numbered list, a table, bullet points, a single paragraph, a 280-character tweet. If you don't specify, AI will guess—and it might guess wrong.

Example: "Format as a numbered list with one sentence per item."

5. Constraints — Set boundaries

Tell AI what NOT to do: "Don't use jargon," "Keep it under 100 words," "Avoid mentioning competitors," "Don't use emojis." Constraints are just as important as instructions.

Example: "Keep the tone professional. No slang. Under 150 words."

Before & After

See the difference good prompts make

Task: Writing an email

Before: "Write an email to my boss."

After: "Write a professional but friendly email to my manager Sarah, requesting a 1-on-1 meeting to discuss my quarterly goals. Keep it under 4 sentences. Suggest meeting this Thursday or Friday afternoon."

Task: Getting recipe ideas

Before: "Give me dinner ideas."

After: "Suggest 5 quick dinner ideas I can make in 30 minutes or less. I have chicken, rice, broccoli, and basic pantry staples. My family doesn't like spicy food. Include estimated prep time for each."

Task: Learning a concept

Before: "Explain blockchain."

After: "Explain blockchain to me like I'm 12 years old. Use a real-world analogy. Keep it under 100 words and avoid any technical jargon."

Level Up

Techniques that take your prompts further

Chain your prompts

Don't try to do everything in one prompt. Start broad, then follow up with specific refinements. "Write a blog outline" → "Now expand section 2" → "Make the tone more casual."

Give examples

Show AI what you want by including an example: "Write a product description like this one: [paste example]. Now write one for my product." AI learns patterns from examples fast.

Ask AI to ask you questions

Try: "I want to plan a vacation. Before giving suggestions, ask me 5 questions to understand my preferences." This lets AI gather the context it needs before answering.

Cheat Sheet

Quick reference: Prompt starters

"Act as a [role]..."
Sets the AI's persona. Great for getting specialized advice or matching a specific writing style.
"Explain [topic] like I'm..."
Controls the complexity level. "Like I'm 5" for ultra-simple. "Like I'm a college student" for more depth.
"Give me [number] options for..."
Forces AI to generate multiple alternatives so you can pick the best one instead of accepting the first idea.
"Before answering, ask me..."
Invites AI to gather context first. Results in much more personalized and accurate responses.
"Rewrite this to be more..."
Great for refining output. "More professional," "more casual," "shorter," "more persuasive."

Try It Out

Test what you've learned with this interactive challenge

Practice makes perfect

The best way to learn prompt engineering is to experiment. Try different approaches, see what works, and refine your technique over time.

Try the Prompt Analyzer Browse More Guides