Why must training and serving use the same feature transforms?
PICTURE THIS: DATA SPLIT
Fit on train, tune on val, report on test once.
Simple meaning
If train uses log(amount + 1) and serve uses raw amount, the model sees a different world and quality collapses.
WHY — Feature Pipelines instead of guessing?
Why interviewers care about Feature Pipelines:
on Feature Pipelines.
the situation, the default choice, and one exception - that reads as experience.
Name the idea, why it exists, then one short example.
End with when you use it and one common pitfall.
STEPS — What happens step by step?
Before you speak the answer, walk the interviewer through these steps:
- 1If train uses log(amount
+ 1) and serve uses raw amount, the model sees a different world and quality collapses.
- 2Why it exists
That mismatch is training-serving skew.
- 3Shared code or a
feature store is how you keep them aligned.
- 4Give an example
One tiny concrete case you can say aloud.
- 5Common mistake
What juniors usually get wrong.
- 6Close
When you pick this over the alternative.
EXAMPLE — See it in action
Here's a short line you can speak, broken into clear beats:
Note: Adapt this scaffold to your own project — keep it under 60–90 seconds.
Key takeaway
If train uses log(amount + 1) and serve uses raw amount, the model sees a different world and quality collapses. That mismatch is training-serving skew.