Capstone lab — evaluation pipeline build

It is 3 AM. The monitoring script will not run until 6 AM. Users have been hitting a broken system for 13 hours and nobody knows. The PM finds out from a VP's angry Slack message the next morning. This actually happens. More often than expected. In L5.6, a monitoring workflow was built that tracks the right metrics and applies the right thresholds. Everything about it is correct. The problem is speed. It runs once a day. And when something breaks at 2 PM, no one finds out until tomorrow morning. This lesson fixes that. Not by replacing what was built; everything from L5.6 still works. The fix is making it faster. Real-time dashboards instead of morning reports. Instant alerts instead of waiting for the next batch run. Statistical drift detection running continuously instead of once a day. This is an advanced optional lab. If L5.6 is complete, the monitoring system is already fully functional. This lesson adds operational speed to that foundation.