504 Gateway Timeout errors and elevated API response times

Update timeline

1. investigating May 11, 2026, 06:50 PM UTC

   Our engineering team has been actively investigating the intermittent 504 Gateway Timeout errors and elevated API response times observed over the past several days. We have identified contributing factors related to how our platform scales capacity during periods of elevated request volume, and we are rolling out staged infrastructure improvements — including load balancer and auto-scaling adjustments — to reduce the likelihood of recurrence during peak load. Service remains operational, and we are monitoring closely. We will follow up here once the initial round of changes is in place.

2. monitoring May 12, 2026, 01:13 PM UTC

   Update: Changes were implemented last night related to the previously identified issues and performance over the past several hours has shown improvement. We will continue to closely monitor system performance, particularly during morning scale-up and evening scale-down periods, to ensure stability.

3. resolved May 13, 2026, 02:00 PM UTC

   We identified the primary cause of the recent instability and 504 errors as related to a core messaging component within our platform. Our team has implemented adjustments to both code and infrastructure, and the issue is currently resolved. We have observed some intermittent spikes during periods when the system scales up or down (typically overnight), but these are minor and not indicative of ongoing instability. We are continuing to monitor performance closely and have initiated planning for longer-term improvements to further enhance stability and scalability.