Walk me through facilitating the post-incident review after a severe outage. How do you keep it blameless but not toothless?

Facilitation is mechanics: timeline before theories, blame converted into system questions, actions with owners tracked to closure. Include the defensive engineer and the exec who wants a head.

Grounded in researched DevOps, SRE and platform engineering interview loops, written to a senior-engineer editorial bar, and never padded to hit a word count.