Observability engineering has emerged as a critical discipline to ensure system reliability, scalability, and health in complex, cloud-native environments, with Prometheus playing a vital role in this ecosystem. Prometheus, an open-source telemetry database, facilitates the collection and storage of custom software metrics and is widely used alongside Kubernetes. To achieve full-stack observability, especially with AI-based platforms like Dynatrace, Prometheus metrics can be ingested and transformed using the OpenTelemetry Collector. This process involves fine-tuning configurations to drop irrelevant labels and metrics, converting cumulative metrics to delta values, and managing memory to optimize performance and reduce storage costs. The use case of Kepler on Kubernetes demonstrates how Prometheus can estimate energy consumption, showcasing the granularity and flexibility offered by observability tools. The Collector's ability to integrate with various data sources and perform necessary adjustments makes it an invaluable asset for managing and transforming telemetry data efficiently before it reaches platforms like Dynatrace, ensuring detailed analysis and actionable insights.