Worked on the GoogleCloudPlatform/monitoring-dashboard-samples repository to enhance the reliability of monitoring dashboards by addressing a critical bug in Prometheus node CPU core count queries. Focused on improving data quality for long time-range metrics, the work involved refining PromQL expressions and updating JSON-based dashboard configurations to ensure accurate capacity representation in GKE environments. This fix reduced the risk of misleading alerts and improved the foundation for capacity planning. Demonstrated proficiency in Kubernetes, Prometheus, and data visualization, while following Git-based workflows to deliver targeted improvements that strengthened trust in monitoring data and supported more informed operational decision-making for dashboard users.