ML foundations
The vocabulary and judgement layer. These questions come up regardless of whether the role is classical ML or LLM-focused, because they’re about how you think rather than what you’ve memorised.
| # | File | The question it answers |
|---|---|---|
| 01 | 01_ml_problem_types.md | turning “we want AI for X” into a problem with a metric |
| 02 | 02_train_val_test_split.md | why the split is where production failures are born |
| 03 | 03_bias_variance_tradeoff.md | “my model isn’t good enough” — diagnosed, not guessed |
| 04 | 04_overfitting_regularization.md | L1 vs L2, dropout, early stopping, and what each actually does |
| 05 | 05_ml_lifecycle.md | end to end, including the unglamorous parts that decide success |
| 06 | 06_when_not_to_use_ml.md | the answer that signals seniority fastest |
File 03 is the highest-leverage one: the train-vs-validation-error table answers most “what do I do next” questions without guesswork.