Developed a reusable Airflow DAG workflow pattern for the HPEEzmeral/aie-tutorials repository, targeting deployments in disconnected or air-gapped environments. The solution leveraged Python and Airflow’s PythonVirtualenvOperator to enable dynamic pip installation with configurable network settings, such as PyPI index URLs, trusted hosts, and proxy support. This approach addressed the challenges of reproducibility and reliability for customers operating in restricted network conditions. By templating the operator and providing traceable change records, the work enhanced auditability and deployment flexibility. The focus on data engineering and DevOps practices ensured that the workflow could be adapted for various isolated deployment scenarios.