Common Prompting Patterns

These proven patterns will help you get better results from AI in various situations.

1. Few-Shot Learning

Show the AI examples of what you want:

Convert these sentences to formal language:

Casual: "Hey, can you send me that file?"
Formal: "Would you please forward the document at your earliest convenience?"

Casual: "Thanks a bunch!"
Formal: "Thank you for your assistance."

Casual: "I can't make it to the meeting tomorrow"
Formal: [AI completes this]

3-5 examples usually work best.

2. Chain of Thought

Ask the AI to think step-by-step:

Solve this problem step by step:

A store has 50 apples. They sell 30% in the morning and 40% of what's left in the afternoon. How many apples remain?

Think through each step before giving the final answer.

This improves accuracy on complex reasoning tasks.

3. Self-Consistency

Get multiple perspectives, then synthesize:

I need to decide whether to learn Python or JavaScript first.

First, argue the case for Python.
Then, argue the case for JavaScript.
Finally, give your balanced recommendation based on both perspectives.

4. Tree of Thought

Explore multiple solution paths:

I want to improve my website's performance.

List 3 different approaches:
1. Quick wins (implement today)
2. Medium-term improvements (this month)
3. Long-term optimizations (this quarter)

For each approach, list specific actions.

5. Persona Pattern

Create consistent character behavior:

You are Alex, a senior software architect with 15 years of experience. You:
- Prioritize maintainability over cleverness
- Always consider scalability
- Prefer simple solutions when possible
- Give direct, honest feedback

Review this code architecture decision...

6. Template Pattern

Use fill-in-the-blank structures:

Generate a product description using this template:

[Product Name] is a [category] designed for [target audience].
Key features include [feature 1], [feature 2], and [feature 3].
Unlike [competitor/alternative], it [unique value proposition].
Perfect for anyone who [use case].

Product: Wireless noise-canceling earbuds

Build on previous outputs:

Draft 1: Write a tagline for a fitness app
[AI responds]

Draft 2: Make it more energetic and action-oriented
[AI refines]

Draft 3: Now make it appeal to busy professionals
[Final refinement]

Quick Pattern Selection Guide

Goal	Pattern
Show examples	Few-Shot
Complex reasoning	Chain of Thought
Decision making	Self-Consistency
Multiple options	Tree of Thought
Consistent voice	Persona
Structured output	Template
Polish results	Iterative Refinement

Combining Patterns

The most effective prompts often combine multiple patterns. For example: Few-Shot + Chain of Thought + Template can produce highly accurate, well-structured outputs.

:::