Splitting 90-day conversation queries to avoid 413 errors

What’s the cleanest way to chunk a 90d date range before hitting POST /api/v2/analytics/conversations/queries? We keep getting 413 Entity Too Large when dateFrom and dateTo span three months. The JSON payload pushes past the limit even with groupBy: ["interval"] set to "15m".

We’ve tried slicing the range and merging page results, but tAnswer aggregates drift when intervals cross month boundaries. Should we just loop through dateFrom increments using analyticsApi.conversationsQuery or is there a hidden maxDays parameter? The worker node timeouts are already maxed out.