Struggling to understand why the Architect HTTP request node consistently returns a 504 Gateway Timeout when invoking our AppFoundry integration endpoint during peak PST business hours. The integration utilizes the Genesys Cloud Platform API v2.5 SDK and handles complex data enrichment for a high-volume queue. The backend service is hosted on AWS US-West-2, which should provide sub-100ms latency to the Genesys Cloud edge nodes. However, under load exceeding 500 concurrent interactions, the timeout threshold appears to be hit regardless of the configured 30-second limit. We have verified that the AppFoundry application is registered correctly with multi-org OAuth scopes and that no rate limiting is active on the partner side. The issue seems to stem from the connection pooling behavior within the Architect engine rather than the application logic itself. Is there a known limitation or configuration parameter for the HTTP request node that manages concurrent connections per flow execution context? We need to determine if this is a platform-side constraint or if specific headers are required to maintain persistent connections during high-throughput scenarios.