Can anyone explain why the Conversational AI analytics endpoint is failing specifically for calls routed through our BYOC trunks?
We are seeing HTTP 500 Internal Server Error responses when querying the /api/v2/analytics/conversations/details endpoint with filters applied to our fifteen BYOC trunks in AP-Southeast-1. The issue appears isolated to conversations that engage the custom bot flow after SIP registration via our primary carrier. Standard PSTN trunks return data correctly within the expected latency window. The error payload indicates a backend timeout during the aggregation of sentiment analysis metrics, which suggests a mismatch in how the bot session data is being correlated with the SIP call legs. We have verified that the trunk registration state remains active and that outbound routing policies are correctly assigning the bot skills group. The problem persists across multiple API client implementations using SDK version 2.4.1. We are attempting to generate a unified quality score report that combines human agent interactions with bot-handled segments, but this data gap is preventing accurate aggregation. Any insights into known compatibility issues between BYOC trunk metadata and the bot analytics pipeline would be appreciated.