Worked on the PrefectHQ/prefect repository to enhance the Prefect Events Client with resilient event handling capabilities. Developed a feature that enables automatic reconnection and resending of unconfirmed events at time-based checkpoints, addressing the risk of data loss in event-driven pipelines. Leveraged Python and Asyncio to implement robust error handling, ensuring events are not dropped during mid-batch resend operations or intermittent network failures. Expanded the test suite to validate reconnection behavior and event persistence under failure scenarios, focusing on backend development and comprehensive testing. This work improved reliability for users operating in environments with unstable network connectivity and demanding data integrity.