Data Validator

Overview

Perform comprehensive data quality checks on datasets — validate schemas, detect anomalies, find duplicates, and enforce data contracts. Essential for ETL pipelines where bad data silently corrupts downstream analytics and dashboards.

Instructions

1. Profile the dataset first

Before validating, understand the data:

• Row count and column count
• Data types per column (string, integer, float, date, boolean)
• Null rates per column
• Unique value counts and cardinality
• Min/max/mean for numeric columns
• Date ranges for temporal columns