Looking for advice on handling 500 Internal Server Error responses from /api/v2/screenrecordings/export when running load tests. The environment is JMeter 5.4.1 targeting Genesys Cloud US1 region. The test plan simulates 15 concurrent agents requesting screen recording metadata exports simultaneously. Under low concurrency (5 threads), the API returns 200 OK with valid job IDs. However, increasing the thread count to 15 causes intermittent 500 errors on approximately 30% of requests. The request payload includes standard fields like start_time, end_time, and agent_id. No 429 throttling is observed, suggesting the issue is not rate limiting but rather backend processing capacity or connection pool exhaustion on the Genesys side. The error response body is minimal, containing only "message": "An unexpected error occurred". Is there a known limit on concurrent export jobs per organization, or should the JMeter configuration be adjusted to stagger requests more aggressively? Any insights into backend capacity constraints for screen recording exports would be helpful for capacity planning.