Our infrastructure is experiencing high latency leading to some instability in core functionality in the Roboflow Platform. Our engineering team is investigating.

We have correlated the elevated latency with an active upstream incident in Google Cloud Platform causing elevated packet loss and errors across multiple products. Our engineering team is continuing to monitor and investigate workarounds.

The underlying systems appear to have stabilized approximately 10 minutes ago and our monitoring dashboards are showing our systems are recovering.

Status changed to resolved.

Google published a public incident page with additional information: https://status.cloud.google.com/incidents/J5ia5t9p3g9Q5Wi7r8Ev