We have now resolved this incident. Between ~7:30AM PT and ~9:30AM PT, a new query pattern caused database slowness that increased API latency and led to dropped inbound calls with certain providers. We mitigated by rolling back the deployment and are following up with a deeper review of the query change.

We are observing systems recover and performance is returning to normal. Some users may still see brief residual impact as systems stabilize.

We are investigating reports of degraded performance affecting a subset of calls. Our team is actively working to determine the root cause and will provide updates as we learn more.