Skip to content
ML Academy

Forecastingdraft — in review

Interview question

  • stafflevel
  • 40 minsuggested time
  • 2Data scientist, ML engineer

Whiteboard case. A retailer wants one forecasting platform for 50,000 SKU-store series (daily, 28 days ahead) that feeds replenishment, with category managers entitled to override numbers. Many series are intermittent; new products have no history. Design the platform: models, hierarchy, evaluation, compute, overrides and governance.

Tests: Large-scale forecasting system design: model portfolio, reconciliation, intermittent demand and cold start, forecast value added, compute budgeting and human overrides.

day 26day 23day 25day 28day 58

Sign in to practise: answers are saved before model answers are shown.