Domain-driven Data Science
summary
Argues domain expertise materially improves outcomes by showing concrete work in supply chain: builds a planning product that combines probabilistic forecasts (exponential smoothing, Holt‑Winters, Croston), operational‑research constraints, and vertical data‑product teams; prototypes interactive tooling with Streamlit prototypes to surface data‑generation quirks; develops a new cold‑start model that reduced error vs baseline but fails adoption because regional processes and trust expect human estimates; ties forecasts to service targets (typical 90% service target) and shows the end‑to‑end costs and tradeoffs.
AI-written summary · about
chapters (7)
- 0:00Domain-driven data science
- 6:11Domain expertise in data science
- 13:23Supply chain logistics
- 20:13Supply chain forecasting
- 35:53Forecast evaluation challenges
- 38:00Real-world domain insights
- 42:32Learning domains and hype
related talks
54mFrom the Life of a Data Scientist
41mData Science – The New Troublemaker in the Delivery Pipeline
37mBridging the gap: from Data Science to service
55mBuilding The Data Dream Team
19mData science demystified
59mDevOps for Data Science
31mAn incomplete list of implementing data science effectively
43mAgile Strategies for High-Impact Data Science & Machine Learning Projects