What belongs in a training config file rather than hidden in code?
PICTURE THIS: DATA SPLIT
Fit on train, tune on val, report on test once.
Simple meaning
Data paths, feature lists, hyperparameters, seeds, and compute settings should be declarative YAML or Hydra configs.
WHY — Reproducibility instead of guessing?
Why interviewers care about Reproducibility:
who only read docs from people who shipped.
and tied to MLOps work.
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:
- 1Data paths, feature lists,
hyperparameters, seeds, and compute settings should be declarative YAML or Hydra configs.
- 2Why it exists
Code should stay generic.
- 3Reviewers can then see
the experiment in the PR diff.
- 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
Data paths, feature lists, hyperparameters, seeds, and compute settings should be declarative YAML or Hydra configs. Code should stay generic.