{"slug":"second-dinner-optimizes-in-game-experiences-with-personalization-on-databricks","url":"https://findausecase.com/use-cases/second-dinner-optimizes-in-game-experiences-with-personalization-on-databricks","title":"Second Dinner optimizes in-game experiences with personalization on Databricks","description":"Marvel Snap developer Second Dinner built real-time, one-to-one shop personalization models using Databricks MLflow and Delta Lake, cutting feature launches from 10 per year to more than 50 (a 400% increase) and reducing model development and training time from months to weeks.","company":"Second Dinner","industry":"Media & Entertainment","aiCapabilities":["Recommendation & Personalization","Machine Learning"],"technology":["Databricks Data + AI Platform","MLflow","Delta Lake"],"deployment":"Unknown","problemStatement":"Second Dinner relied on manual, heuristic-driven processes to collect player data and generate dashboards; a single feature iteration could take a month, limiting the team to 10 personalization use cases yearly, and personalization lagged behind player interest since it was based on aggregate statistics rather than true personalization.","solutionApproach":"Second Dinner built an ML model in Databricks using MLflow, ingesting customer engagement data through Delta Lake to feed the models, and used PyFunk (a light wrapper from MLflow) to let data users add logic via Python; pre-calculated shop recommendations are now globally deployed using a low-latency, model-serving endpoint.","businessValue":"Second Dinner increased feature launches 400%, from 10 per year to more than 50, and cut model development and training time from months to weeks, enabling real-time, one-to-one shop personalization for every active player globally.","evidence":{"band":"high"},"sourceUrl":"https://www.databricks.com/customers/second-dinner","dates":{"publishedAt":"2026-09-15T09:05:18.014Z","publishedAtSource":"pipeline","updatedAt":"2026-09-15T09:05:18.014Z"},"license":"Open for reading and citing with a link to https://findausecase.com/use-cases/second-dinner-optimizes-in-game-experiences-with-personalization-on-databricks. Bulk republication requires permission."}