Kuyun AI · Works to try

Feature and data pipelines

Skew, backfills, feature stores and data quality, five rounds against the clock

practice · by @interview_room

For a senior machine learning engineer or ML platform engineer preparing for a loop where the questions are about the data under the model: how features are computed, kept consistent between training and serving, backfilled without leaking the future, and watched for silent breakage. Each round is a 45-minute mock interview with a different interviewer who runs the clock as a real one would.

What's inside · 5

  1. Training-serving skew
  2. Backfills and point-in-time correctness
  3. Feature stores
  4. Data quality monitoring
  5. Design: features for fraud detection

Sign in, or sign up if you are new — either way you land on this work.