AI Prompt Engineering Cheat Sheet
A quick-reference guide to the prompting patterns and techniques worth using, for people who’d rather get work done than read theory about it.
Why Prompt Engineering Matters
The difference between a mediocre AI response and a brilliant one is almost always the prompt. Prompt engineering is the skill of crafting inputs that guide AI models to produce accurate, relevant, and useful outputs. It’s one of the most valuable skills for anyone working with AI tools today.
You don’t need to be technical. The best prompt engineers are clear thinkers who know how to give good instructions, a skill you already use every day.
The 7 Core Principles
Be Specific
Vague prompts produce vague results. Include details about format, length, tone, and audience. The more specific your input, the more useful the output.
Assign a Role
Tell the AI who it should be: “You are an experienced HR manager” or “Act as a data analyst.” This frames the response with the right expertise.
Provide Context
Give background information the AI needs. Include relevant data, constraints, and the situation. Context turns generic answers into tailored solutions.
Use Examples
Show the AI what you want with one or two examples of ideal output. This is called “few-shot prompting” and dramatically improves consistency.
Break It Down
Complex tasks should be split into steps. Ask the AI to “think step by step” or chain multiple prompts together for better reasoning.
Set Constraints
Define boundaries: “Keep it under 200 words,” “Use bullet points,” “Write for a non-technical audience.” Constraints focus the output.
Iterate & Refine
Treat your first prompt as a rough draft. Read the output, identify what’s missing, and refine. Great results come from conversation.
Prompt Template
Use this universal template as your starting point for any task:
RoleYou are a [role/expert type] with expertise in [domain]. Task[Clear description of what you want the AI to do] Context[Background information, relevant data, or situation details] Format[How you want the output structured: bullet points, table, paragraph, etc.] Constraints[Word limit, tone, audience, what to avoid] Example (optional)[An example of what ideal output looks like]
Quick-Reference Techniques
Zero-Shot Prompting
Ask the AI to perform a task without providing any examples. Works best for straightforward tasks where the model has strong base knowledge.
// Zero-shot: No examples needed Prompt: “Classify the following review as Positive, Negative, or Neutral: ‘The product arrived on time but the packaging was damaged.’”
Few-Shot Prompting
Provide 2–3 examples of the input/output pattern you want. The AI learns the pattern and applies it to new inputs.
// Few-shot: Show the pattern Review: “Amazing quality, will buy again!” → Positive Review: “Broke after two days.” → Negative Review: “It works as described, nothing special.” → Neutral Review: “The product arrived on time but the packaging was damaged.” →
Chain-of-Thought (CoT)
Ask the AI to explain its reasoning step by step. This dramatically improves accuracy for complex or multi-step problems.
Simply adding “Let’s think step by step” to the end of your prompt can noticeably improve accuracy on complex, multi-step tasks.
Common Pitfalls to Avoid
- Being too vague: “Write something about marketing” vs. “Write a 300-word LinkedIn post about B2B email marketing trends for SaaS companies.”
- Overloading a single prompt: If you need multiple outputs, split them into separate prompts.
- Ignoring the output: Always review and refine what AI gives you before you use it.
- Not providing format guidance: If you want a table, say so. If you want bullet points, ask for them.
- Forgetting audience: A prompt for a C-suite executive should produce very different output than one for an intern.