ETL pipeline that reconciles partner billing with internal subscription data
Finance teams need a reliable view of revenue versus entitlements, and this project delivers a daily ETL that matches transactions, flags mismatches, and classifies discrepancies. It uses a pure‑Python reference engine plus optional PySpark/Delta on Databricks, with ADF orchestration and a medallion layout on ADLS for reproducible runs. The pipeline is aimed at finance and product engineers who need drill‑downable, auditable reconciliation reports, and it stands out by being runnable locally without a Spark cluster while still supporting full enterprise deployment.
View on GitHub →nomi96/Tapmad_Case_Study