The Observability Data Platform (ODP) powers the core of Datadog's telemetry systems, handling exabytes of multimodal observability data. As AI agents become first-class consumers of telemetry, ODP is evolving to meet their demands - scaling with explosive data growth, exposing new query mechanisms, rethinking how telemetry is stored, transformed, and served, and enforcing guardrails that ensure security and reliability. Our team's new focus is to build an intelligent control plane for production systems. This involves moving beyond passive monitoring to create a platform where AI agents can safely and effectively take action in live environments. To achieve this, we are integrating techniques from symbolic reasoning, formal methods, and generative AI. We are looking for an experienced Staff Applied Scientist with a background that spans systems engineering, AI, and formal reasoning. You have expertise in areas like causal modeling, generative simulation, runtime verification, or reinforcement learning, and are motivated to apply these skills to build reliable systems. You will join the team behind Datadog's most ambitious projects: evolving observability infrastructure for stochastic, self-improving systems.