The kinds of things DF kids will roast you for...

- Using ChatGPT (arguably okay if used responsibly)
- Unused import statements
- Saying not to look at part of the code
- "We did test it"... but it's not in your notebook
- "I found the middle ground"... how?
- No F1 score
- "You're hiding your low precision"
- Computing accuracy score manually instead of using a library
- Using a decision tree instead of a random forest (aathreya: eh)
- "You guys just have accuracy and it's printed manually"
- "What are you going to do in that exact same interview where they ask about class imbalence?"
- No hyperparameters
- "Hella simple model"
- Uninterpretable models