Managing observability across 49 cloud accounts (AWS, Azure, GCP, Digital Ocean), 1900+ servers, 21K+ Kubernetes pods (AKS, EKS, GKE), and hundreds of applications presents unique challenges. This session shares practical methodologies and tooling for instrumenting, collecting, and correlating metrics, logs, and traces in such a massive, diverse environment. Learn how we leverage open-source and CNCF projects alongside tools like Kong, Sentry, SonarQube, and Kafka to achieve meaningful alerting, insightful dashboards, and proactive performance management for our OTT platform. Discover strategies for unified querying, effective alerting policies, and managing observability at hyper-scale.