We all have that one super intelligent friend who knows almost everything- history, science, literature, politics, you name it. They may not be your friend, but at least you know someone who fits that bill. Or you might even be that person.
We Nigerians call people like this efiko. They can somehow answer any question under the sun. The thing with many people like this is that they’ll only give you the best answer if you ask the right question. I know someone who would throw your question back at you and would only respond when you’ve framed your question clearly and ‘intelligently’. That’s prompt engineering in a nutshell.
The Hype About Prompt Engineering
When you ask your favorite AI assistant to write a poem, craft a story, solve a math problem, or plan your next vacation, you’re engaging in the art of prompt engineering, whether you realize it or not. But what exactly is prompt engineering, and why is everyone talking about it? Let’s get right into it!
What is Prompt Engineering
In the world of artificial intelligence (AI), prompt engineering refers to designing specific, clear, and purposeful instructions known as prompts, to guide AI systems like ChatGPT or Gemini to produce the results you want. It’s like giving a recipe to a chef: the better the recipe, the better the dish!
And why does this matter? Because AI isn’t a psychic who can read all your thoughts. You may be brainstorming ideas for your office strategy session, coding solutions, or searching for content ideas, how you phrase your request can significantly impact the response. With AI becoming a core part of our tech-driven lives, mastering this skill can make you a star in both professional and personal spaces. Alexa, play me Megan Thee Stallion’s “Watashi wa star”.
Understanding Prompts: The Secret to This AI Thing
At its core, a prompt is just an input or a question you type into an AI system. For example:
“Write a motivational email to my team.”
“Summarize this 10-page document into bullet points.”
“Tell me a dad joke about robots.”
How Clear Prompts Help AI Serve Up the Perfect Response
AI models like ChatGPT use these prompts, treating them like clues. They’ve been trained in a huge amount of information- think of it as AI reading every book in the local library, reading every online article and reading every newspaper that’s been published! When you type something in, the AI uses that knowledge to guess the best and most relevant answer based on what you’ve asked. It’s like solving a puzzle: the clearer your clue (prompt), the better the AI can piece together the perfect response. Get it?
So, here’s the thing: if your prompt is vague or ambiguous, the AI might deliver ‘low budget’ results. Imagine that you’re about to cook for your family- you ask them what they’d like to eat, they rack their brains trying to think up the best dishes, then end up saying, “um, I dunno, just make something nice.” You might end up making a steamy pot of goat meat pepper soup or you could as well make them a simple slice of toast bread. Their vagueness really leaves you no choice. 🤷🏼♀️ The same applies to AI.
To get top-tier results, you need to be:
- Clear: Use simple, precise language.
- Contextual: Provide background or specifics.
- Concise: Avoid overloading with unnecessary details.
Now, let’s get into real-world examples featuring ‘The Good, The Bad, and The Ugly’ prompts. (Only legends know this classic movie…don’t ask me about it, ask ChatGPT).
Prompt Crafting 101: What Works and What Doesn’t
THE BAD | THE UGLY | THE GOOD |
SCENARIO 1 – WRITING BLOGS | ||
“Write a blog.” AI’s Response: “Sure, but what’s the topic? Length? Audience? Help me out here!” (You’ll spend more time clarifying than getting useful results.) | “Write a blog on something cool about AI.” AI’s Response: “AI is cool because it’s everywhere!” (Super generic and not particularly useful- clearly, AI needs more direction!) | “Write a blog for beginners explaining SEO, including its definition, SEO Optimization, and why it’s important.” AI’s Response: “Absolutely! Here’s a detailed blog draft for you…” (AI loves it when you hit the nail on the head!) |
SCENARIO 2 – CODING | ||
“Help me with coding.” AI’s Response: “What kind of help? Debugging? Learning? Please specify!” (AI is not a mind reader!) | “Fix my code; it’s broken.” AI’s Response: “Please provide the code, the error message, and what you’re trying to achieve.” (You’ve given AI almost nothing to work with, and now you’re stuck in a Q&A loop!) | “I’m working on a Python project. Can you explain how to use loops to iterate through a list of numbers?” AI’s Response: “Sure! Here’s how loops work in Python, with an example…” (Clear, specific, and solves your exact problem.) |
SCENARIO 3 – PLANNING A VACATION | ||
“I want to go on vacation. Help.” AI’s Response: “Where do you want to go? What’s your budget? Can you give me more details?” (Frustratingly vague, right? You’re not getting far with this one.) | “Plan me a vacation somewhere cool.” AI’s Response: “How about visiting Antarctica for an icy adventure?” (Technically “cool”, but this suggestion might freeze your budget- or your entire body!) | “I’m looking for a budget-friendly beach vacation in Africa during December. Can you suggest destinations and activities?” AI’s Response: “Sure! You can visit Zanzibar for its beautiful beaches and affordable accommodation. Activities include snorkeling, visiting Stone Town, and sunset dhow cruises.” (A solid response that helps you plan your dream trip!) |
SCENARIO 4 – WRITING A SOCIAL MEDIA CAPTION | ||
“Write me a caption.” AI’s Response: “What kind of caption? For what platform? What’s the vibe?” (You’ll end up doing more explaining than you wanted.) | “Write a funny caption for my photo.” AI’s Response: “Sure: ‘Want to hear a joke? Me, that’s it!’ “ (Uh…now, why would you want to throw shades at yourself?) | “Write a witty Instagram caption for a photo of me holding a coffee cup on a rainy day.” AI’s Response: “Rainy days and coffee are always a vibe because caffeine spurs my COFFEdence!” (Short, witty, relatable, and perfect for Instagram!) |
SCENARIO 5 – LEARNING TO COOK | ||
“Teach me to cook.” AI’s Response: “What do you want to cook? What’s your skill level? Do you have ingredients?” (Too many follow-up questions—you’ll lose interest before you start.) | “I want to cook something fancy for dinner.” AI’s Response: “Try making Casu Marzu. It’s an impressive dish!” (You may want to ask AI the meaning of Casu Marzu. You’ll probably end up ordering takeout instead!) | “I’m new to cooking and want to make a tasty Nigerian dish. Can you give me an easy recipe with step-by-step instructions?” AI’s Response: “Absolutely! Here’s a recipe for jollof rice: simple, delicious, and beginner friendly.” (A clear response tailored to your needs.) |
When it comes to working with AI, the more specific your prompts, the better the results. Avoid being vague or overloading the AI with too much information. Be clear, add relevant details, and avoid leaving too much room for interpretation. AI and your future self will thank you!
Conclusion: Why Mastering Prompt Engineering Matters
In a world where AI is quickly becoming your co-worker, study buddy, research assistant and sometimes even your therapist, prompt engineering is the skill to have. It helps you:
- Get precise, helpful results.
- Save time by avoiding back-and-forth clarifications.
- Discover the full potential of AI tools.
Think of prompt engineering as learning a new language, a language that helps you communicate better with AI. The good news is that it’s fun, easy to master, and endlessly rewarding. Move away, Duo (no beef intended, Duolingo fans), but this reader has got some language to learn. 🤨
So, go ahead- experiment, tweak, and refine your prompts. Before long, you’ll be the go-to person for “making AI work smarter, not harder.” Happy prompting!
PS: In the coming weeks, we will be launching a stellar course on Prompt Engineering that can help you achieve so much more as a professional. So, here is a prompt for you:
Head over to goverticul.io and sign up to get first-hand info and a sweet discount when the course drops.