/Molly Zhu

How We Built Infrastructure to Run User Forecasts At Spotify tl;dr: "To enable fast iterations and facilitate model training, we built a system that allows user forecasts to run both on demand and automatically every week, with the heavy hyperparameter tuning part running on the weekend. The system creates visualizations and automated insights to surface information, and provides easy-to-use tools for manual interventions for both data scientists and business stakeholders."

featured in #332