Enterprise Retail & supply chain
Retail Demand Forecasting Platform
Lead Data Scientist · H-E-B
A 22-repository GCP platform forecasting SKU-store sales and promotional redemption for ~350 stores over an 89-week horizon.
- lower WMAPE vs. legacy
- ~50%
- pipeline runtime
- −72%
- stores, 89-week horizon
- ~350
- on-demand promo forecast
- 25–30 s
The problem
Planners needed SKU-by-store forecasts for regular and promoted items far enough ahead to plan buys and promotions, and the legacy model was both less accurate and slow.
What I built and led
- Led a cross-functional team of data scientists and engineers, setting technical direction and MLOps standards across 22 repositories.
- Built the production models: LightGBM for weekly SKU-store sales and promotional redemption (benchmarked against XGBoost and a Temporal Fusion Transformer) and ARIMA+ for daily demand, trained on Ray in Vertex AI, served via BigQuery ML and fed by Dataform feature pipelines.
- Designed the evaluation and explainability layer — leakage-audited backtests by category, promotion and horizon week, plus SHAP dashboards for planners — and an experimentation platform.
- Delivered an on-demand Cloud Run forecasting API returning per-promotion forecasts in about 25–30 seconds.
- Earlier, led modeling for the legacy promotional forecasting platform this replaced, and helped refactor it from R and Scala to PySpark on AWS, cutting its runtime by nearly 70%.
Results
- ~50% lower WMAPE vs. legacy
- −72% pipeline runtime
- ~350 stores, 89-week horizon
- 25–30 s on-demand promo forecast