During January 2025, Slabe enhanced observability for the AI-Hypercomputer/JetStream repository by developing a new Prometheus gauge metric to monitor model load times. Using Python and Prometheus, Slabe updated the backend server library to measure and report the duration required to load engine parameters, enabling more precise startup diagnostics. The implementation included adding a sanitized 'model_name' label to metrics, allowing for granular, per-model performance analytics and improved differentiation across deployments. This work deepened JetStream’s monitoring capabilities, supporting more effective performance tuning and reliability analysis. The feature was delivered through two focused commits, reflecting a targeted and well-scoped engineering effort.