We're experiencing degraded service on our REST API. The engineering team is investigating.

Service is restored and has been stable for the past 10 minutes. We're monitoring & searching for the root cause.

We have identified a bottleneck caused by ephemeral Clickhouse latency as the likely cause of the downtime. We are working to add a new queue between our API & Clickhouse to mitigate the impact of downstream issues on the API layer.