How would you implement environment promotion: dev, staging, prod for both data and models?
PICTURE THIS: DATA SPLIT
Fit on train, tune on val, report on test once.
Simple meaning
Dev uses sampled data and mock stores
WHY — CI/CD for ML instead of guessing?
Why interviewers care about CI/CD for ML:
question about CI/CD for ML.
trade-offs, and what you would actually do on a MLOps project - not buzzwords.
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:
- 1Dev uses sampled data
and mock stores
- 2staging uses production-like schemas
with masked PII and the candidate model
- 3prod uses signed artifacts
and real stores.
- 4Promotion copies immutable versions,
it does not retrain with different untracked data.
- 5Access and quotas tighten
at each stage.
- 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
Dev uses sampled data and mock stores staging uses production-like schemas with masked PII and the candidate model