Streaming Analytics

This project implements an end-to-end analytics platform for a Netflix or Spotify-style streaming media service. It ingests catalogue (titles), users, sessions, view events, and ratings into Snowflake. dbt transforms them through a medallion architecture into engagement marts, content-performance marts, and a recommender mart. The recommender — the dissertation's centerpiece — is implemented as a dbt Python model executing in Snowpark, using item-based collaborative filtering with implicit-feedback signals.