Whenever you're ready, walk me through how you'd approach this.
What are the key metrics you would monitor for a microservices architecture with Datadog?
Explain how you would organize the metrics across services, dependencies, infrastructure, and user-facing behavior. Describe which Datadog products or features you would use, how you would set alerts and SLOs, and how you would detect correlated failures without creating excessive alert noise.