Vizient Inc.
Automated Validation Pipeline for Quality Metrics
Engineered an end-to-end SAS + Excel validation pipeline that cut manual review time by 40% and hardened accuracy across ambulatory quality metric releases.
The Problem
Validating multi-domain ambulatory quality and accountability metric releases relied on slow, manual review — a bottleneck that risked inconsistencies before metrics reached clients.
Approach
I engineered an end-to-end automated validation pipeline in SAS and Excel, then layered in Python benchmarking scripts that compare extract and result datasets to catch discrepancies proactively.
Background
Vizient publishes ambulatory quality and accountability metrics across multiple domains. Each release had to be validated for accuracy — historically a manual, time-consuming process.
What I Built
I built an end-to-end automated validation pipeline in SAS and Excel that standardized the checks required for every release. To make the process self-auditing, I developed Python benchmarking scripts that compare extract datasets against result datasets and flag any misalignment automatically.
Outcome
- 40% less manual review time per release
- 100% alignment between extract and result datasets
- Proactive, automated detection of data-quality issues before publication
Impact
- Cut manual review time by 40% across metric releases
- Achieved 100% alignment between extract and result datasets
- Enabled proactive detection of data-quality issues
- Improved accuracy and consistency across multi-domain releases