Explore how Intuit tackles P1/P2 outages caused by changes in their Kubernetes environment through a conference talk on observability and streaming machine learning. Learn about the development of Numaflow, a Kubernetes-native DAG-based streaming processing platform, and Numalogic, a streaming ML platform, both open-sourced under Numaproj. Discover how these tools collect, process, and analyze in-cluster data in real-time, computing anomaly scores for each deployment. Gain insights into the integration of observability into Argo CD, enabling users to understand and remediate change-induced behavior. Understand how this approach helps reduce Mean Time to Detect (MTTD) and Mean Time to Resolve (MTTR) in a large-scale Kubernetes environment with approximately 2,500 services.