It's back down again, new incident: https://status.claude.com/incidents/34wq7gxnxvvm
I still wonder if LLMs were used to "resolve" the first incident...
What's there to wonder? The premiere AI company is bound to dogfood their product to help with the main problem space their product is meant to tackle.