Worked on the GoogleCloudPlatform/vertex-ai-samples repository to address a critical issue affecting the reliability of demonstration notebooks for online prediction model monitoring. Focused on correcting resource paths within the model_monitoring_for_custom_model_online_prediction.ipynb notebook, the work ensured that model file locations and BigQuery dataset URIs referenced publicly accessible sample data. By resolving broken links and updating access URLs, the changes eliminated authentication barriers and improved reproducibility for users exploring Vertex AI samples. Leveraged expertise in Python, cloud technologies, and data engineering to streamline the setup process, enabling seamless execution of machine learning demos without additional configuration or authentication requirements.