AI Code Review Checklist¶

For Levels 5 and above. Use this checklist after writing your solution and before moving on.

Before Asking AI to Review¶

Complete these checks yourself first. AI review is most useful when you have already caught the obvious issues.

Does it run without errors?¶

python project.py

If it crashes, fix it yourself first. Do not ask AI to debug a program you have not tried to run.

Do all tests pass?¶

pytest -q

If tests fail, read the failure output. Try to fix it based on what the test expected vs what your code produced. Only ask AI for help after you have attempted a fix.

Do I understand every line?¶

Go through your code line by line. For each line, ask yourself: "What does this do? Why is it here? What would happen if I removed it?"

If you cannot answer those questions for any line, that is the line you need to study, not skip.

Have I tested edge cases?¶

Think about inputs that are unusual or extreme: - Empty input (empty string, empty list, empty file) - Very large input - Invalid input (wrong type, missing fields, negative numbers) - Boundary values (zero, one, the maximum)

Try at least two edge cases that the provided tests do not cover. Write them down in notes.md.

Is it idiomatic Python?¶

Idiomatic means "the way experienced Python programmers would write it." Some questions to ask: - Am I using list comprehensions where a simple loop would be clearer? Or vice versa? - Am I using with open() for file operations (not bare open())? - Am I using f-strings instead of string concatenation? - Am I using pathlib.Path instead of string manipulation for file paths? - Are my variable names descriptive and following snake_case?

Could I explain this to someone else?¶

Imagine a fellow learner asks you how your code works. Could you walk them through it without looking at the code? If not, you do not understand it well enough yet.

What would I change if I had more time?¶

Write one or two things you would improve in notes.md. This builds the habit of critical self-review that professionals use every day. Examples: - "I would add better error messages" - "I would split this long function into two smaller ones" - "I would add more tests for the edge case where the file is empty"

Asking AI for Code Review¶

Once you have completed the checklist above, share your code with AI using this format:

I wrote this Python code for [project name]. It passes all tests.
Here is my code:

[paste your code]

Please review it for:
1. Code quality and readability
2. Edge cases I might have missed
3. More idiomatic Python alternatives
4. Any potential bugs

Do not rewrite it for me. Point out specific lines and explain what could be improved.

Key phrase: "Do not rewrite it for me." You want feedback, not a replacement. Rewriting it yourself after getting feedback is where the learning happens.

After AI Review¶

Read each suggestion carefully
For each suggestion, decide: do I agree? Why or why not?
Implement the changes yourself (do not copy AI rewrites)
Run tests again to make sure your changes did not break anything
Update notes.md with what you learned from the review

Quick Reference¶

Step	Check
1	Runs without errors
2	All tests pass
3	I understand every line
4	Edge cases tested
5	Idiomatic Python
6	Could explain it aloud
7	Know what I would improve

Home