operational resilience
Railway founder berates “beyond insane” GCP move to cut off API, VMs, database instance with no warning
Agentic development workloads and use of large monorepos stretch GitHub to breaking point
Regulators, CISOs alike remain concerned about concentration risk
Notable by their absence?
"The goal is shifting from automating recovery to sustaining continuity. When a SaaS outage occurs, resilience depends not on faster failover, but on clear processes, tested workarounds, and business-led decisions"
"We have identified the cause and are rolling out changes to restore normal service"
"If this had been flag protected, the issue would have been caught in staging"
Too much sunshine, not enough inertia?