Introduction
An AI coding assistant generates code from the prompts you give it. The output often looks reasonable, but it can contain problems that aren't immediately visible. AI can generate code that:
- Calls functions that don't exist
- Works, but is inefficient with large sets of data
- Misses edge cases
- Has security vulnerabilities
Code review is the process of reading code carefully before you use it. AI-generated code requires particular attention because it can contain confident but incorrect output. If you understand how code should work, you can identify many of these problems.
This module teaches you to:
- Identify incorrect information in AI-generated code.
- Find inefficiencies that slow code down as data grows.
- Identify security risks, such as injection, hardcoded secrets, and unsafe file paths.
- Apply a structured code review process.
- Decide whether AI-generated code is ready for production.
The goal isn't to distrust every output. It's to verify that the code works before anyone relies on it.