Study Notes

Appearance

Domain 5: Improve Developer Productivity with GitHub Copilot (10–15%) ​

← Domain 4 · Next Domain →

Exam Tip

This domain is practical — the exam presents real-world scenarios and asks how Copilot should be used to solve them. Know the right Copilot feature for each task: inline suggestions for generation, Chat for explanation, /tests for testing, /fix for debugging.

Code Generation, Refactoring, and Documentation ​

Code Generation ​

Copilot generates code from:

  • Comments: Describe what you want in a comment → Copilot fills in the implementation
  • Function signatures: Write the function name and parameters → Copilot generates the body
  • Open files context: Copilot infers patterns from existing code in your project

Best uses: Boilerplate code, repetitive patterns, scaffolding new functions/classes, generating CRUD operations

Refactoring ​

Use Copilot Chat for refactoring tasks:

/fix — Fix a specific issue in selected code
"Refactor this function to use async/await" — general refactoring
"Convert this class to use dependency injection" — architectural refactoring

Agent Mode or Edit Mode for cross-file refactoring (e.g., renaming an interface that's used in 10 files)

Documentation Generation ​

/doc — Generate inline documentation for selected code
"Write a README for this module" — via Copilot Chat
"Add JSDoc comments to all exported functions" — Agent Mode

Reducing Context Switching ​

One of Copilot's highest-value use cases is keeping developers in flow:

  • Learn as you code: Ask Copilot to explain unfamiliar APIs or libraries without leaving the IDE
  • Generate sample data: "Give me 5 sample JSON objects for a user profile" — no need to switch to a browser
  • Modernize legacy code: Ask Copilot to translate Python 2 → Python 3, or convert callbacks to async/await
  • Shell command help: Use Copilot CLI instead of searching Stack Overflow for CLI syntax

Testing with GitHub Copilot ​

Generating Unit Tests ​

python
# Select your function, then use:
# /tests in Copilot Chat
# or write a comment: "# Test that get_email returns None for invalid input"

Copilot can generate:

  • Happy path tests (valid inputs)
  • Edge cases (empty strings, None, boundary values)
  • Error cases (exception handling)

Identifying Edge Cases ​

Ask Copilot Chat: "What edge cases should I test for this function?"

Copilot will enumerate cases like:

  • Empty input / null
  • Very large inputs
  • Boundary conditions (e.g., exactly at the limit)
  • Invalid types
  • Concurrent execution scenarios

Writing Assertions ​

python
# Copilot can generate assertion-level tests:
# "Write assertions to verify the returned dict has keys 'id', 'email', 'role'"

Security Improvements and Performance Optimizations ​

Security-Aware Code Review ​

Use Copilot Chat to review code for security issues:

"Review this function for SQL injection vulnerabilities"
"Is this code vulnerable to XSS?"
"Suggest a more secure way to store this password"

Performance Optimizations ​

"Rewrite this loop using vectorized NumPy operations"
"Optimize this SQL query to avoid N+1 selects"
"Replace this O(n²) algorithm with an O(n log n) approach"

Trap

Copilot can suggest security and performance improvements but cannot guarantee they are correct. Always test and review AI-generated security changes with a security expert.

Domain 5 Quick Quiz

1 / 4
❓

Which Copilot Chat command generates unit tests for selected code?

(Click to reveal)
💡
/tests — generates unit tests for the currently selected code block.

← Domain 4 · Next Domain →