A production release last week (week of 6/24) introduced a change that negatively impacted performance when loading the application. For at least 2 users this caused the application to fail to load at least once. This slow performance was detected by our automated monitoring over the weekend and our engineers investigated the issue but did not identify the root cause. This (Monday) morning the root cause was determine fix was deployed by 6:52am.
We've started a RCA and post-mortem, which includes adding automated checks on the performance of the key API endpoints that were impacted.