Developed robust data quality tolerance parameters for equality and non-equality checks in the databrickslabs/dqx repository, enhancing the reliability of numeric data validation. Leveraging Python and Spark, the work introduced absolute and relative tolerance options to data quality functions, allowing for more flexible comparisons and reducing false positives in analytics pipelines. The implementation included comprehensive unit, integration, and manual tests, as well as practical usage examples and documentation to support maintainability. Collaboration with other contributors ensured alignment with data quality standards and readiness for release, establishing a foundation for more resilient and confident data validation across diverse workloads.