Learn Prompt Engineering

Core ideas for new AI users — in workshops and at work.

Prompt Engineering Best Practices

1. Be Specific and Clear

Vague prompts lead to vague results. Clearly define what you want.

❌ Bad:

"Write about marketing"

✅ Good:

"Write a 500-word article about digital marketing strategies for small businesses, focusing on social media and email campaigns."

2. Provide Context

Give background information to help the AI understand your needs better.

Example:

"[Context: You are helping a startup founder] Create a business plan for a SaaS product targeting small businesses..."

3. Use Structured Format

Organize your prompt with clear sections using brackets and labels.

Structure:

[Role]
[Context]
[Task]
[Output Format]
[Constraints]

4. Specify Output Format

Tell the AI exactly how you want the response formatted.

"Format your response as: 1) Executive summary, 2) Key points (bullet list), 3) Action items (numbered list)"

5. Include Examples (Few-Shot)

Show the AI what you want by providing examples of desired outputs.

"Example 1: Input: 'happy' → Output: 'joyful, cheerful, elated'
Example 2: Input: 'sad' → Output: 'melancholic, sorrowful, downcast'
Now process: 'excited'"

6. Set Constraints and Boundaries

Define what should and should not be included in the response.

"Do not include technical jargon. Keep explanations simple. Maximum 200 words."

7. Iterate and Refine

Use the “Refine Again” feature to continuously improve your prompts based on results.

Prompt Examples Library

📝 Content Writing

Example prompt for blog posts and articles

💼 Business & Strategy

Example prompt for business analysis

🎨 Creative Writing

Example prompt for stories and narratives

🔬 Technical Documentation

Example prompt for technical content

📊 Data Analysis

Example prompt for analyzing data

🎯 Problem Solving

Example prompt using Chain of Thought

Pro Tips & Tricks

💡

Start Simple, Then Expand

Begin with a basic prompt, then add details and constraints iteratively.

🎯

One Task Per Prompt

Focus on a single objective. Multiple tasks can confuse the AI.

📏

Set Length Limits

Specify word count or character limits to control output size.

🔄

Use Iterative Refinement

Refine your prompt multiple times, incorporating feedback from previous outputs.

🎭

Assign Roles

Use role-based prompting: “You are an expert [role]…” for better context.

✅

Test Different Approaches

Try different templates (RTRF, COSTAR, Chain of Thought, etc.) for the same task.

🔍

Be Explicit About Format

Specify JSON, markdown, tables, or other formats explicitly.

🚫

Avoid Ambiguity

Replace vague terms with specific, measurable requirements.

Template Selection Guide

RTRF Framework

Best starting point for beginners: Role, Task, Requirements, Format.

When to use:

Everyday work emails and summaries
Corporate workshop practice
Quick, clear prompts without complexity
Teaching AI basics to new users

COSTAR Framework

Best for complex tasks requiring structured thinking and comprehensive responses.

When to use:

Business strategy and planning
Professional content creation
Multi-faceted problem solving
When context and clarity are critical

Chain of Thought (CoT)

Perfect for breaking down complex problems into step-by-step reasoning.

When to use:

Mathematical problems
Logical reasoning tasks
Multi-step problem solving
When you need to see the thinking process

Few-Shot Learning

Ideal when you have examples of desired input-output patterns.

When to use:

Pattern recognition tasks
Classification problems
Format transformation
When examples clarify expectations

Role-Based Prompting

Use when you want the AI to adopt a specific perspective or expertise.

When to use:

Expert advice and consultation
Educational content
Professional writing
When perspective matters

Structured Output

Best for when you need responses in a specific, consistent format.

When to use:

Data extraction
Formatted reports
Structured analysis
When format consistency is critical

Turn everyday tasks into clear AI prompts

Everyday Prompt Assistant

Your prompt is ready

Why this assistant

Guided for beginners

Workshop frameworks

Quality score

We don’t save your data

Learn Prompt Engineering

Prompt Engineering Best Practices

1. Be Specific and Clear

2. Provide Context

3. Use Structured Format

4. Specify Output Format

5. Include Examples (Few-Shot)

6. Set Constraints and Boundaries

7. Iterate and Refine

Prompt Examples Library

📝 Content Writing

💼 Business & Strategy

🎨 Creative Writing

🔬 Technical Documentation

📊 Data Analysis

🎯 Problem Solving

Pro Tips & Tricks

Start Simple, Then Expand

One Task Per Prompt

Set Length Limits

Use Iterative Refinement

Assign Roles

Test Different Approaches

Be Explicit About Format

Avoid Ambiguity

Template Selection Guide

RTRF Framework

COSTAR Framework

Chain of Thought (CoT)

Few-Shot Learning

Role-Based Prompting

Structured Output

Useful Tools & Resources