Worked on the aws/amazon-cloudwatch-agent repository to enhance CloudWatch integration by refactoring the entity attribute mapping process. Focused on implementing dynamic cluster attribute mapping based on Kubernetes mode, this work improved the accuracy and flexibility of how entity attributes are processed and mapped to CloudWatch Entity fields. Leveraging Go and Kubernetes, the changes reduced mapping errors and increased the maintainability and extensibility of the monitoring pipeline. The enhancements enabled more reliable dashboards and alerts by validating the impact on monitoring fidelity and data correctness across clusters, supporting scalable observability and more robust monitoring workflows in diverse cloud environments.