Dynatrace has launched Dynatrace Intelligence, a system that combines deterministic AI and agentic AI. The platform is designed to help organizations transition from reactive to autonomous operations.
Dynatrace Intelligence is the new agentic operations system that takes center stage at the observability company’s Perform conference. It is built to observe and optimize dynamic AI workloads. The platform is designed to enable organizations to build more resilient applications and improve customer experiences.
The core innovation lies in combining two AI approaches. Deterministic AI provides precise insights based on real-time causal context. Agentic AI, in turn, can safely reason, make decisions, and take actions within defined boundaries. These two techniques are combined within the operating system.
The system runs on the Dynatrace platform. This includes Grail, the unified data lakehouse that stores metrics, logs, traces, events, and business data. It also uses Smartscape, the real-time dependency graph that automatically maps relationships.
Dynatrace benchmarked an external SRE (site reliability engineer) agent working with its deterministic agents. Problems were solved up to twelve times more often, three times faster, and at half the cost compared to tests without deterministic agents.
Specialized Agents for Different Domains
The introduction of Dynatrace Intelligence also includes new domain-specific AI agents for SREs, developers, and security teams. These agents are designed to help organizations translate real-time observability insights into fully autonomous operations. The agents are built on Dynatrace Intelligence. The system works with different types of agents, each fulfilling a specific function. Domain agents coordinate end-to-end outcomes for SRE, DevOps, business observability, and security operations. Assist agents in interpreting situations and guiding users through the best next steps using natural language.
In addition, agents provide fundamental capabilities such as causal reasoning, predictions, and real-time intelligence. Agentic workflows orchestrate complex objectives with policy-driven controls. Finally, ecosystem agent integrations enable Dynatrace Intelligence to collaborate with external systems.
The agents mobilize automatically when triggered by detected anomalies or user requests. They assess the context, determine the urgency, and coordinate next steps. Actions are executed immediately through tools teams already use for communication and work tracking.
Ecosystem of agents
Companies can orchestrate built-in and partner agents. The platform has bidirectional integrations with ServiceNow, AWS, Microsoft Azure, Google Cloud, Atlassian Jira, GitHub, and Red Hat.
The agentic architecture also consists of multiple layers. Agents provide fundamental capabilities for a trusted operational context through causal reasoning and real-time intelligence. Other agents extend teams with insight and guidance for specific functional areas. Ecosystem agents connect to partner platforms and expand the scope of autonomous actions.
From insights to autonomy
With Dynatrace Intelligence, organizations can achieve self-healing systems in dynamic, AI-driven environments. The platform supports proactive prevention, recovery, and optimization. Teams remain in control while the system manages operational complexity in the background.
Dynatrace emphasizes supporting companies in a phased journey to autonomy. Organizations can start with AI-driven insights and recommendations. They can then use automation for controlled operations with human oversight. Ultimately, they can transition to fully autonomous operations with safeguards and controls.
“Agentic AI offers enormous potential, but many businesses still struggle to ensure it operates reliably, securely, and with consistent performance in real‑world environments,” said Bernd Greifeneder, Chief Technology Officer and founder of Dynatrace. “Dynatrace Intelligence fuses deterministic and agentic AI, removing the guesswork and delivering AI‑powered observability organizations can trust.”
Tip: Dynatrace moves from APM and observability to Autonomous Intelligence