Partial-Key and Multi-Column Duplicates

Concepts covered: sqlWindowDedup

The dedup operation produces a count of removed duplicates per run. That count is itself a metric. A pipeline that silently dedupes is a pipeline whose data quality issues are invisible; a pipeline that emits the duplicate count and alerts on drift catches upstream regressions before consumers do. This section covers the metric, the alert, and the operational pattern. The duplicate-count metric Two CTEs compute the source row count and the deduped row count. The INSERT writes a single row into a pipeline_metrics table with all three numbers. Over time, the metric table contains the duplicate count per pipeline run; a dashboard or alert reads from it. The cost is a few extra queries per run; the benefit is observability into the dedup's behavior. 3 The alert pattern Alert on drift: 'the dup

About This Interactive Section

This section is part of the Deduplication: Intermediate lesson on DataDriven, a free data engineering interview prep platform. Each section includes explanations, worked examples, and hands-on code challenges that execute in real time. SQL queries run against a live PostgreSQL database. Python runs in a sandboxed Docker container. Data modeling problems validate against interactive schema canvases. All content is framed around what data engineering interviewers actually test at companies like Meta, Google, Amazon, Netflix, Stripe, and Databricks.

How DataDriven Lessons Work

DataDriven combines four interview rounds (SQL, Python, Data Modeling, Pipeline Architecture) with adaptive difficulty and spaced repetition. Easy problems get harder as you improve. Weak concepts resurface until you master them. Your readiness score tracks progress across every topic interviewers test. Every lesson section ends with problems you solve by writing and running real code, not by picking multiple-choice answers.