The API and user-facing application in the US data center were unresponsive for around 3 minutes. Both services resumed normal operation automatically.

We will investigate a root cause to understand whether any changes are necessary to avoid similar incidents in future.