Saving Private Production

In this Ignite talk, I share a real incident where a seemingly small and safe change turned into a full production outage. I’ll walk through what actually went wrong, why the failure was predictable in hindsight, and how the team managed to recover. The focus is not on blame or tools, but on the patterns behind these incidents: hidden coupling, overconfidence in automation, and changes that quietly cross a critical boundary — a bridge too far.