Worked on the wazuh/wazuh-indexer repository to deliver enhancements for remote cluster state monitoring, focusing on improving observability and reliability in distributed systems. Refactored the backend logic handling remote cluster state downloads and publications, separating statistics for download failures and publication failures to enable more granular fault isolation. Introduced new metrics for incoming publication failures and updated existing metrics to accurately reflect the source of issues, supporting faster diagnosis and targeted remediation. Leveraged Java and system design expertise to enhance monitoring granularity, streamline root-cause analysis, and reduce mean time to recovery for remote state management in production environments.