55 mins lesson duration•11 mins read

Lesson 7: Prompting for Code Generation

Write effective prompts for generating, debugging, and refactoring code across programming languages.

Code Generation Prompts

Getting LLMs to write code effectively requires understanding how to communicate technical requirements clearly. The best code prompts are specific about language, context, and constraints.

The CODE Framework

Letter	Meaning	What to Include
C	Context	What is the project? What does the code need to do?
O	Output	What language, framework, style?
D	Dependencies	What libraries/APIs are available?
E	Examples	Show similar code or expected behavior

Mental model: Think of code prompts like giving specs to a developer — the more specific your requirements, the less revision needed.

Generating New Code

Basic Code Prompt:

Write a Python function that:
- Takes a list of dictionaries with 'name' and 'score' keys
- Returns a new list sorted by score in descending order
- Handles empty lists gracefully
- Includes type hints and docstring

Advanced Code Prompt:

You are a senior Python developer following PEP 8 style guidelines.

Create a function with these specifications:
- Function name: calculate_moving_average
- Input: data (list of floats), window_size (int)
- Output: list of floats (moving averages)
- Edge cases: empty data → empty list, window_size > len(data) → None
- Include: type hints, docstring with examples, error handling

Dependencies: Only use standard library (no numpy/pandas)

Example usage:
>>> calculate_moving_average([1, 2, 3, 4, 5], 3)
[2.0, 3.0, 4.0]

Key elements for code generation:

Specify the programming language explicitly
Describe inputs, outputs, and edge cases
Mention style preferences (PEP 8, etc.)
Include example usage
Specify what libraries are available

Debugging Code Prompts

When you have an error:

I'm getting this error in my Python code:

[Error message]

Here's the code:
[Code snippet]

What's causing this error and how do I fix it?

For systematic debugging:

This code should [expected behavior] but instead [actual behavior]:

[Code]

Please:
1. Identify the bug(s)
2. Explain why they occur
3. Provide the corrected code
4. Explain what changed and why

When you don't understand the error:

I don't understand why this code fails:

[Code]
[Error message]

Please:
1. Explain what the error means in plain English
2. Walk through the code execution step-by-step
3. Identify where it goes wrong
4. Provide a fix with explanation

Refactoring Code Prompts

Basic refactoring:

Refactor this code to be:
- More readable
- Following [language] best practices
- Better modularized

Here's the current code:
[Code]

Keep the same functionality but improve the structure.

Performance optimization:

This function works but is too slow for large datasets:

[Code]

Optimize it for:
- Better time complexity
- Memory efficiency
- Maintaining readability

Explain what optimizations you applied.

Code review style:

Review this code as a senior developer would:

[Code]

Provide feedback on:
1. Correctness
2. Performance
3. Readability
4. Security concerns
5. Best practices violations

Then provide an improved version.

Testing Code Prompts

Unit test generation:

Write unit tests for this function:

[Function code]

Requirements:
- Use pytest framework
- Cover normal cases, edge cases, and error cases
- Include descriptive test names
- Aim for 90%+ code coverage

Test case generation:

For this function: [function signature]

Generate test cases that cover:
1. Happy path (expected inputs)
2. Boundary values (min, max, empty)
3. Error conditions (invalid inputs)
4. Edge cases specific to the logic

Present as a table with: Test Name | Input | Expected Output | Description

Code Explanation Prompts

For learning:

Explain this code to a junior developer:

[Code]

Break it down:
1. What does this code do at a high level?
2. Line-by-line explanation
3. Key concepts used
4. Potential improvements
5. Common mistakes to avoid

For documentation:

Generate comprehensive documentation for this code:

[Code]

Include:
1. Overview description
2. Function/class documentation (JSDoc/docstring format)
3. Parameter descriptions
4. Return value documentation
5. Usage examples
6. Known limitations

Language-Specific Considerations

Language	Key Prompt Details
Python	Mention PEP 8, type hints, virtual environments
JavaScript/TypeScript	Specify ES6+, async/await, module system
React	Hooks vs class components, state management approach
SQL	Database type, schema context, query optimization
Go	Error handling patterns, goroutines if applicable
Rust	Ownership model, unsafe blocks if needed

Common Mistakes to Avoid

Mistake: Vague requirements like "make it better" — Fix: Specify exactly what improvements you want.
Mistake: Not providing context about existing codebase — Fix: Share relevant code structure and conventions.
Mistake: Forgetting to mention constraints — Fix: Always specify language version, available libraries, etc.
Mistake: Not testing generated code — Fix: Always run and verify code before using it.

Professional Tips & Tricks

For complex functions, break into smaller prompts (one per component).
Always test generated code in a safe environment first.
Include error handling requirements upfront — don't add it later.
Use "show your reasoning" for complex algorithms.

Key Takeaways

Use the CODE framework: Context, Output, Dependencies, Examples.
Be explicit about language, style, and constraints.
For debugging, include both the code and the error message.
For refactoring, specify what aspects to improve.
Always test generated code before using it in production.

Next up: Prompting for content creation and writing tasks.

Interactive Lesson Code Snippet

# Code Prompt Examples

## Example 1: Function Generation
Prompt: "Write a Python function to validate email addresses using regex"

Generated Code:
import re
from typing import Optional

def validate_email(email: str) -> bool:
    """
    Validate an email address using regex.
    
    Args:
        email: The email address to validate
        
    Returns:
        True if valid, False otherwise
        
    Examples:
        >>> validate_email("user@example.com")
        True
        >>> validate_email("invalid-email")
        False
    """
    pattern = r'^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}$'
    return bool(re.match(pattern, email))

## Example 2: Debugging Prompt
Prompt: "This code crashes with IndexError. Fix it."

Code:
def get_middle(lst):
    return lst[len(lst) // 2]

Fix: Add empty list check
def get_middle(lst):
    if not lst:
        return None
    return lst[len(lst) // 2]

Language: python

Lesson Code (Python)

# Code Prompt Examples

## Example 1: Function Generation
Prompt: "Write a Python function to validate email addresses using regex"

Generated Code:
import re
from typing import Optional

def validate_email(email: str) -> bool:
    """
    Validate an email address using regex.
    
    Args:
        email: The email address to validate
        
    Returns:
        True if valid, False otherwise
        
    Examples:
        >>> validate_email("user@example.com")
        True
        >>> validate_email("invalid-email")
        False
    """
    pattern = r'^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}$'
    return bool(re.match(pattern, email))

## Example 2: Debugging Prompt
Prompt: "This code crashes with IndexError. Fix it."

Code:
def get_middle(lst):
    return lst[len(lst) // 2]

Fix: Add empty list check
def get_middle(lst):
    if not lst:
        return None
    return lst[len(lst) // 2]

Console Output

Code Prompt Examples

## Example 1: Function Generation
Prompt: "Write a Python function to validate email addresses using regex"

Generated Code:
import re
from typing import Optional

def validate_email(email: str) -> bool:
    """
    Validate an email address using regex.
    
    Args:
        email: The email address to validate
        
    Returns:
        True if valid, False otherwise
        
    Examples:
        >>> validate_email("user@example.com")
        True
        >>> validate_email("invalid-email")
        False
    """
    pattern = r'^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}$'
    return bool(re.match(pattern, email))

## Example 2: Debugging Prompt
Prompt: "This code crashes with IndexError. Fix it."

Code:
def get_middle(lst):
    return lst[len(lst) // 2]

Fix: Add empty list check
def get_middle(lst):
    if not lst:
        return None
    return lst[len(lst) // 2]

Code Visualization Tips

Create a code prompt checklist with all required elements.
Draw a flowchart for the debugging prompt process.
Compare good vs bad code prompts side by side.

Professional Tips & Tricks

For complex code, break into multiple prompts — one per function.
Always specify the language version and available libraries.
Include example inputs and outputs for clarity.

Python Code Judge & Practice Arena

LeetCode Style

Run real Python 3.12 WebAssembly code directly in your browser against automated test suites.

Solved:0 / 2

0 / 30 XP

Challenges:

Problem 1 of 2

Code Generation Prompt

Medium+20 XP

Write a detailed prompt that would generate a Python class for managing a todo list with add, remove, complete, and list operations.

main.pyPython 3.12 (WASM)

# Code Generation Prompt
# Write your solution below

def solve():
    pass

Press Run Code to test or Submit to verify test cases

Test Your Knowledge

Instant feedback

Quick Check: Prompting for Code Generation

1 / 2

What does the CODE framework stand for?

Up next · Continue learning

Prompting for Content & Writing

Master prompts for blog posts, marketing copy, technical writing, and creative content.

10 mins read50 mins

Start next lesson

Previous: Error Recovery & Edge Case Handling Next: Prompting for Content & Writing

Made withbyAmol Shukla·amolshukla.online