Developed a new onFlush callback for the bulk helper in the elastic/elasticsearch-js repository, enabling users to access cumulative statistics such as documents processed, successes, and failures after each flush. This feature enhanced observability and troubleshooting for bulk operations by surfacing real-time metrics, supporting more reliable integrations with the Elasticsearch JS client. The implementation involved designing an API consistent with existing bulk operations, writing comprehensive unit tests, and updating documentation to guide users on the new functionality. Work was carried out using JavaScript and TypeScript, with a focus on Node.js full stack development and robust testing practices throughout the process.