With Intelligence, Dynatrace focuses on fully automated IT operations, deeply rooted in observability. The AI agent can autonomously detect and resolve problems.
At its annual Perform conference, Dynatrace announced Dynatrace Intelligence: an AI-driven agent system that helps companies manage IT environments autonomously and securely. The announcement was supported by various innovations in observability, cloud integrations, and developer tools.
With Dynatrace Intelligence, the company is taking the next step towards what it calls ‘autonomous IT operations,’ where companies can manage their digital infrastructure with more confidence and less manual work.
From detection to remediation
Dynatrace Intelligence combines deterministic AI with agentic AI, which can make decisions within controlled parameters and based on causal context. This combination enables companies to perform autonomous actions faster and more reliably. This helps them detect incidents faster, mitigate risks, and proactively optimize systems.
Dynatrace is building on existing observability capabilities of its platform such as Smartscape and Grail, the data lakehouse where all data is stored. “AI capabilities were already embedded in the platform for problem detection,” says Chris Geebelen, Regional Director Solutions Engineering at Dynatrace, to ITdaily. “Now we are linking it with concrete actions, so that remediation becomes possible without human intervention. This saves time and reduces the business impact of problems.”
The system works with internal and external agents, including integrations with AWS, Azure, Google Cloud, ServiceNow, and GitHub.
Broader ecosystem strengthens new functions
Dynatrace Intelligence also includes domain-specific agents. These support SRE, development, and security teams in automatically detecting and resolving problems. The agents work together via ‘agentic workflows,’ which allow companies to automate more complex goals with policies and approvals. External platform integrations ensure that actions are performed within an organization’s existing tools.
For developers, there are new tools that link observability data to real-time debugging, error analysis, and front-end problems. Thanks to the acquisition of DevCycle, developers can also use feature-level runtime controls for safer release management. New integrations strengthen ties with the major hyperscalers AWS, Microsoft Azure and Google Cloud.
In addition, Dynatrace launched an enhanced version of Real User Monitoring (RUM), which combines front-end data with back-end context. This helps teams improve user experiences in complex, AI-driven applications. “Giving AI all the necessary context leads to better answers. There’s nothing worse than having AI perform actions on incorrect data,” Geebelen concludes.
There’s nothing worse than having AI perform actions on incorrect data
Chris Geebelen, Regional Director Solutions Engineering Dynatrace