Developed a reliability enhancement for ETL pipelines in the astronomer/airflow repository by introducing the BigQueryStreamingBufferEmptySensor. This feature detects when the BigQuery streaming buffer is empty before allowing DML operations on streaming tables, addressing the risk of unprocessed data affecting downstream results. The solution leverages Python and integrates with Airflow sensors to automate checks within cloud-based ETL workflows. By focusing on data engineering best practices and utilizing BigQuery’s streaming capabilities, the work improved end-to-end data correctness and reduced runtime errors. The contribution demonstrates a methodical approach to ETL development, emphasizing reliability and correctness in cloud computing environments.