Kaizen! Letting our new CDN instances crash so we can learn and improve | Gerhard Lazu

Why watch

This Changelog episode covers Gerhard Lazu's Kaizen approach to improving infrastructure by intentionally allowing new CDN instances to crash in order to analyze failure modes. The discussion focuses on debugging out-of-memory errors, reviewing Pipedream instance status checkers, and applying systematic operational improvements. Targeted at DevOps engineers and infrastructure developers interested in learning from production incidents.