spot-check-features

What problem does it solve?

This Skill helps data teams detect and validate data quality issues in the ML feature store before model training, reducing the risk of training on dirty data.

Core Features & Use Cases

• Feature store quality checks: validates readiness and alert levels across recent data slices.
• Comprehensive metrics: tracks per-date quality metrics such as matchup_quality_pct, vegas_quality_pct, and feature_count completeness.
• Operational validation: provides SQL templates and dashboards to identify data quality regressions and guide remediation.

Quick Start

Run the validation against the last 7 days of feature data to surface quality readiness and alert levels. Example queries can be executed with your data warehouse client to reproduce checks for is_quality_ready, quality_alert_level, and feature_count completeness, and to surface actionable remediation guidance.