Data platform teams using Amazon Redshift collect performance telemetry from system views and Amazon CloudWatch metrics, but correlating issues like a spike in QueryRuntimeBreakdown commit time with small INSERT statements or high disk spill with undersized compute requires deep expertise. A new AI-powered solution collects this telemetry, pre-computes performance signals, correlates them with CloudWatch, and uses Amazon Bedrock to generate prioritized recommendations. The solution uses a signal-based design where a collector Lambda runs 13 diagnostic SQL queries against Amazon Redshift Serverless, reads the workgroup’s WLM configuration, and collects CloudWatch metrics across capacity, query execution, WLM, connections, and storage, then writes a telemetry JSON file to Amazon S3. An analyzer Lambda reads the telemetry, builds a structured prompt with inline CloudWatch-to-signal correlations, calls Amazon Bedrock (Anthropic Claude Sonnet 4.6), and writes the recommendations JSON back to Amazon S3. An Amazon SNS topic sends an email summary of top recommendations. Prerequisites include an Amazon Redshift Serverless workgroup, a superuser database administrator, and Amazon Bedrock model access.