Developed and integrated Prometheus SLA Target Monitoring for the ai-dynamo/dynamo repository, focusing on enhancing observability and reliability of planning workflows. Implemented gauge metrics to track time-to-first-token and inter-token latency, wiring these metrics into the planner’s initialization process to surface real-time SLA performance data. Leveraged Python for backend development, utilizing Prometheus for monitoring and unit testing to ensure robust metric reporting. This work enabled the exposure of SLA target metrics to Prometheus, supporting data-driven decision making and providing actionable business insights through real-time dashboards. The feature improved transparency into planner performance, aligning with reliability and optimization objectives for the system.