sdecoret - stock.adobe.com
Dynatrace acquires Arize for AI agent development
With its third acquisition this year, Dynatrace continued its march from pure observability to systems of control, adding AI agent development tools to its observability arsenal.
As enterprise AI maturity grows, Dynatrace is betting that demand for custom AI agent development will, too, and is anticipating that demand with a $915 million deal to acquire a venture-backed startup.
The cash-and-stock deal for Arize AI, a 200-employee company based in Berkeley, Calif., and founded in 2020, will add its AI agent evaluation and experimentation tools to Dynatrace's portfolio, where they will be combined with the recently unveiled Bluebox AI product. Bluebox automates the regular software development lifecycle (SDLC); the Arize AX platform can perform similar autonomous improvement cycles on custom AI agents.
"Arize was built as an agent-first platform, with one of the largest collections of agent trace data in the industry," wrote Arize co-founder and CEO Jason Lopatecki in a blog post. "Dynatrace brings depth in tracing and logging software systems that provide the data to drive these agents. These two areas just belong together if we are going to build the future systems that continuously improve."
In other words, Dynatrace has runtime telemetry; Arize can monitor, revise and refine the output quality of large language models; and Bluebox can then fix the agents based on both tools' findings, said Torsten Volk, an analyst at Omdia, a division of Informa TechTarget.
"The AI observability [Dynatrace] already had was just focused on standard metrics, not on actually looking at model outputs in different constellations," Volk said.
One Dynatrace customer considering Bluebox said he is intrigued by the addition of Arize to support custom agent development alongside other types of software.
"DIY agents are now like websites: We're gonna have millions of them, and most of them are untested and will not scale," wrote Mark Tomlinson, AVP of performance and observability at digital payments provider FreedomPay, in an email to TechTarget. "Bluebox + Arize is a nice combination that bolsters Dynatrace's long-standing commitment to 'find and fix' value delivery."
This acquisition follows the January 2026 acquisition of feature flagging specialist DevCycle, which established Dynatrace's strategy to move more aggressively into what chief technology strategist Alois Reitbauer described as "active systems of control." In April 2026, Dynatrace bought data pipeline startup Bindplane.
Enterprises want unified software, agent development
Dynatrace isn't the only enterprise IT vendor looking to expand into AI agent development. In April, Cisco acquired Galileo, which performs AI agent evaluations and integrates with Splunk observability tools. Harness added support for AI agent development to its existing DevSecOps pipelines with Agent DLC in July.
Analysts said all of these vendors are responding to a growing desire among enterprises for a single platform to develop AI agents and other software applications. An Omdia survey of 400 IT professionals and application developers, conducted in late 2025 and published in March, found that "Ability to monitor AI/ML systems with the same observability platform used for infrastructure and applications" ranked second on respondents' priority list, with 49% rating it important. Only core monitoring capabilities, such as accuracy, drift, latency and cost, were rated important by more respondents, at 53%.
While it's still early for enterprise AI agent development, "AI agent adoption is only going to explode upwards," said Stephen Elliot, an analyst at IDC.
"Agent debugging, evaluations, agent root cause analysis and testing … might even be more important than they were in the traditional SDLC model, especially with agentic systems that make decisions," Elliot said. "I view it as a foundational investment for technology executives."
