Prompting 101: The Crash Course I Wish I Had

I bet you've experienced this too.

I once asked ChatGPT to write a course curriculum—with an “inspired” prompt.

It absolutely nailed the tone.

With each retry, the output got better and better. It made me think "Wow, this is exactly what I wanted!"

Then a couple of days later, I wrote a prompt for the same task, but the outputs were boring, generic, and flat—just predictable, average content you could find with a quick Google search.

Same AI. Almost identical prompt. Just a few words different.

I think right now at a stage of the "AI hype cycle" where what I'm going to say is just evidence — I will shock no one:

Prompting is the most important and high-leverage skill right now.

IT'S JUST THE NEW LITERACY. LIKE LEARNING BASIC MATH 100 YEARS AGO!

Prompting isn't about knowing what to ask. It's about how you ask it.

The model doesn’t make the magic.

You do.

This newsletter is a crash course in prompt engineering—a collection of proven frameworks, plus guidance on fixing bad prompts (using SUSHI 🍣).

🧭 Why Bother With Prompting in 2025?

Prompt literacy is the Google-fu of this decade: trivial to ignore, priceless when mastered.

🧱 Prompting 101 — The 5-Ingredient Framework

Level 1: Foundation.

Here’s the basic prompt recipe. Use it every time.

Mnemonic: Tall Cats Read Every Issue

(Task, Context, References, Evaluate, Iterate)

1. Task

What should the AI do? Be painfully specific.

✅ Good:

“Explain 3 evidence-backed memory strategies. Present in a table: name, summary, effort level, evidence grade. You’re a cognitive scientist.”

❌ Bad:

“Help me study.”

2. Context

Who is this for? What’s the end goal?

✅ Good:

“Write for product engineers in their 40s learning to learn. Tone: clear, witty, no jargon. Goal: make cognitive science feel useful.”

3. References

Want better tone or structure? Show it.

✅ Example:

“Imitate the tone of Make It Stick: ‘Spaced practice feels harder, but results in better retention. Effort now, payoff later.’”

4. Evaluate

This step is often overlooked.

• Use Perplexity or another LLM to double-check output.
• Compare to a real blog post.
• Ask: Does this hit the mark?

5. Iterate

Tweak. Rerun. Improve.

Prompting is like sculpting.

Your first pass is just the block of marble.

🍣 Level 2: Iteration — Sushi Solves Most Constraints

Prompt not working? Don’t panic. Reframe it using the Sushi method:

S.U.S.H.I. = Simplify, Shift Perspective.

Then people added; Modify Language, Impose Constraints.

🔹 Simplify

Break it down. Avoid prompt soup.

❌ “Design a homepage with brand values + unique selling points + CTAs.”

✅ “Describe brand-aligned homepage. Add CTA section above fold.”

🔹 Shift Perspective

Change the role to change the result.

❌ “Act as a researcher. Summarize this paper.”

✅ “Act as a Wired journalist. Pull out one surprising insight. Use storytelling.”

🔹 Modify Language

Tone matters. Ask better, not louder.

❌ “Make this sound better.”

✅ “Rewrite in Brené Brown’s voice — warm, vulnerable, grounded.”

🔹 Impose Constraints

Boundaries = creativity.

❌ “Suggest book titles.”

✅ “Suggest 5 speculative fiction titles under 5 words, using alliteration.”

🚀 Level 3: Expert Moves (AKA Prompting Like a Pro)

You’re past beginner. Let’s go advanced.

1. Zero-Shot Prompting

How to use: give clear instructions without providing examples.

When to use: when the task is straightforward, such as translations or factual queries.

PROMPT: Translate the English phrase 'Flowers on the road' to Spanish.

2. One-Shot Prompting

How to use: give clear instructions along with a single example to demonstrate the task or desired output.

When to use: when a single example is sufficient to clarify the task or output format for the model.

PROMPT: In uppercase, return the Spanish trasnlation of the English word 'basket' only.

EXAMPLE:
English word (input): River
Spanish translation (output): RÍO

3. Few-Shot Prompting:

How to use: Provide clear instructions with multiple examples in the prompt.

When to use: When you need to adapt the model to a specific task or domain without fine-tuning, and you want more consistent and accurate outputs.

PROMPT: return the sentiment of this statement 'The lecture was quite boring' only. Either positive, negative, or neutral.

EXAMPLES:
'This movie was great!': Positive,
'I hated the service.': Negative,
'I don't know how I feel about it.': Neutral

4. Role Prompting

How to use: Give clear instructions and assign a specific persona to the model to shape its responses.

When to use: When the task is open-ended and the output needs to align with a specific perspective, personality, or tone.

PROMPT: Write a short blog (500 words) with 4 points about college hacks.
‍
ROLE: Act as a sweet college girl who uses a lot of Gen z slangs.

5. Style Prompting

How to use: Include specific style, tone, or genre requirements in the prompt.

When to use: When you need the output to match a specific style or tone.

PROMPT: Write a brief formal email requesting a raise.

6. Emotion Prompting

How to use: Include emotionally resonant language or sentiment in your prompt to guide the tone and feeling of the response.

When to use: When generating creative content like stories and poetry that needs emotional depth and authenticity.

PROMPT: Write a poem about my lost imaginary friend who never gave up. I still miss my friend.

7. Contextual Prompting

How to use: provide relevant background or situational information before stating clear instructions in the prompt.

When to use: When background context or domain-specific details are needed to make the response more accurate and relevant, such as in RAG chatbots.

CONTEXT: My name is Jennifer Luke and I'm a marketing manager in JL firm.
‍
PROMPT: Write an email to the team about the upcoming campaign.

8. Rephrase and Respond (RaR)

How to use: Ask the LLM to first rephrase your question into an optimized prompt, then generate the answer based on that improved version.

When to use: For complex queries where you want to verify the LLM's understanding and get more accurate responses.

PROMPT: Rephrase and expand the following question, and then answer it: What is the difference between correlation and causation?

9. Re-reading (RE2)

How to use: After stating your initial instruction or question, add "Read that again:" and repeat the original instruction/question verbatim.

When to use: For complex tasks that involve reasoning.

PROMPT:
A farmer has a rectangular field that is 3 times as long as it is wide. The perimeter of the field is 400 meters. What are the dimensions of the field?

Read the question again: "A farmer has a rectangular field that is 3 times as long as it is wide. The perimeter of the field is 400 meters. What are the dimensions of the field?"

10. System Prompting

How to use: Provide high-level instructions or context that will guide the LLM throughout your entire interaction.

In ChatGPT, you can do this using the 'customize GPT' feature.

When building LLM apps, you can set this through the 'system prompt' in your configuration.

When to use: When you need to establish consistent behavior and tone for the LLM in a conversation.

SYSTEM PROMPT: You are a helpful assistant that will provide factual responses in a concise tone

11. Self-Ask

How to use: Ask the LLM to break the question into smaller sub-questions and have it answer each one to arrive at the final answer.

When to use: When the task is complex, requires reasoning, or involves multiple steps.

PROMPT:
Should I pursue a master's degree in data science?
Break this question into smaller sub-questions, answer them, and provide a final recommendation based on your reasoning.

12. Chain-of-Thought (CoT)

How to use: Ask the model a question and add the phrase "Let's think step by step"

When to use: For tasks that require reasoning, such as math or logic problems.

PROMPT: What is the total cost of a meal with a 10% discount and a 7% tax?
Let's think step by step.

13. Step-back Prompting

How to use: Ask a broad, foundational question first, then prompt the model to address a specific question based on its broader understanding.

When to use: When the task requires analysis or decision-making that needs to consider multiple underlying factors and principles.

PROMPT: Explain the key factors that influence a company's decision to expand into a new market.
Based on this, should a company in the tech industry expand to Europe?

14. Self-Consistency

How to use: Ask the LLM to generate multiple responses to the same prompt and select the most frequently occurring answer.

When to use: When dealing with tasks that have multiple possible solutions and require high consistency and accuracy.

PROMPT: What is the most popular programming language for machine learning?
Generate 5 possible answers and return only the one that appears most often.

15. Thread-of-Thought (ThoT)

How to use: Similar to Chain-of-Thought, but instead of saying "let's think step by step," you'll say "walk me through this in manageable parts step by step."

When to use: When the task involves question-answering and large complex contexts, such as in RAG systems.

CONTEXT: I have a problem involving a group of people attending a party. There are 10 guests: Alice, Bob, Carol, Dave, Eve, Frank, Grace, Henry, Irene, and Jack. Each person has a preference for the type of music they want to listen to, and there are 3 types of music: Jazz, Rock, and Classical. Each person will only attend if they get to hear their preferred type of music. However, due to limited speakers, only 3 music genres can be played, and only one genre can be played at a time.
‍
PROMPT: Walk me through this in manageable parts step by step to figure out the maximum number of guests who can attend the party ( the maximum number of people who can be satisfied with their music preference).

16. Tree-of-Thought (ToT)

problem into smaller steps. At each step, the model should explore multiple possible solutions, evaluate their merit, and proceed with the most promising option until reaching a final solution.

When to use: When the task requires deep reasoning, multi-step planning, and high accuracy.

PROMPT: I am designing a new type of coffee cup that keeps drinks hot longer.

Break down this problem into smaller steps. At each step, generate multiple possible solutions, evaluate their quality (consider factors like feasibility, cost-effectiveness, and potential impact), and continue with the best option until you reach a final solution.

Start by brainstorming initial design concepts.

17. ReAct (Reason and Act)

Take action based on that thought, observe the outcome, and use these observations to refine future thoughts and actions.

When to use: When the task requires iterative decision-making and interaction with external systems or data.

PROMPT: I need to find the latest market trends for electric vehicles.
First, generate a thought about the most relevant keywords to search for.
Then, perform a search using those keywords by calling the search API.
Observe the results of the search, refine the keywords based on the data retrieved, and conduct another search if necessary.
Repeat this process, adjusting your strategy based on the latest findings, until you find the most relevant and recent market trend data

🧪 Research-Proven: Prompting Boosts Performance

🎯 Final Dare: Write the Worst Prompt You Can

Do it.

Seriously.

Open your favorite model.

Write the worst, vaguest, most useless prompt you can think of.

Now improve it five times.

By round five, you’ll be better than 90% of LinkedIn’s “AI experts.”

You’ll be amazed how far one good prompt can take you.

Happy prompting.

—Charafeddine