Dynatrace completes acquisition of Arize to advance full-lifecycle AI observability

Dynatrace has completed its acquisition of Arize. The acquisition combines Arize’s AI-native tracing, evaluation, and experimentation capabilities with Dynatrace observability across applications, infrastructure, user experiences, and business processes. Together, these capabilities create the foundation for full-lifecycle AI observability – from development and evaluation through deployment, operation at scale, and continuous improvement.

Arize builds directly for AI engineers, with tracing, evaluation, experimentation, and observability workflows that help teams understand and improve AI applications and agentic systems. Teams can inspect agent trajectories, model calls, retrieval, tool use, context, performance, and cost; run evaluations and compare experiments; and use what they learn to improve the next iteration. Dynatrace serves the SRE and platform engineering teams responsible for the broader digital environment, providing observability across applications, services, infrastructure, user experiences, and business processes. The acquisition brings these disciplines together to connect how AI applications and agents are built and improved with how they perform and the outcomes they deliver.

When we announced our intent to acquire Arize in August, we described a broader challenge: the disciplines required to build, run, and improve AI have evolved separately. AI engineering teams use traces, evaluations, datasets, and experiments to understand behavior and improve AI applications and agents. SRE and platform engineering teams manage applications, services, and infrastructure around them. Business stakeholders measure the user experiences and outcomes they produce. But the signals and workflows across these teams – and across the AI lifecycle – too often remain disconnected.

As agentic systems take on more complex and consequential work, enterprises need more than visibility. They need a continuous path from development and experimentation through deployment, operation, and improvement, with shared context about agent behavior, system performance, cost, risk, user experience, and business outcomes.

Dynatrace and Arize are building that path.

What does this mean for enterprises scaling AI?

AI now spans the lifecycle from experimentation to customer-facing products and business-critical workflows. As AI applications and agents influence decisions and take action across systems, organizations are accountable not only for whether those systems are available, but for how they behave and what outcomes they produce.

Understanding that behavior starts with the path an agent took. Which models and agents were involved? What context did it retrieve? Which tools did it call? What actions did it take, and what did that path cost? Without this context, inefficient trajectories can lead to unexpected spending, while inappropriate access or actions can create security and compliance risks. Application and infrastructure telemetry remains essential, but it cannot answer these questions on its own.

A harder question is whether the system produced the right outcome. AI can fail silently, producing a confident, plausible response that is wrong without triggering an error. An agent can complete a workflow but still take the wrong path, act on stale context, or update the wrong record. Services can appear healthy even when the outcome is wrong.

And no AI system remains fixed. Updates to models, prompts, retrieval strategies, tools, data, or agent workflows can alter its behavior. Teams need evaluations and experiments throughout the lifecycle – not only before deployment or after something goes wrong. What they learn should become a new dataset, regression test, evaluator, or experiment that helps improve the next iteration.

The signals needed to do this often exist, but they are distributed across different tools, workflows, and teams. AI engineers evaluate behavior and compare experiments. SRE and platform engineering teams understand applications, services, infrastructure, and dependencies. Business and product teams see the effects on users, processes, and outcomes.

Our shared vision is to connect AI engineering workflows with the applications, infrastructure, user experiences, business processes, and outcomes around them. Technical health alone does not determine whether an AI system is working as intended. Teams need context across the lifecycle to understand what happened, identify where an issue originated, assess what it affected, and apply what they learn to the next iteration.

Arize AX and the Dynatrace platform remain available as standalone offerings, and Phoenix remains available as an open-source project. With the acquisition complete, our teams can begin connecting these capabilities. We will share updates as our plans and priorities develop.

What does this mean for AI builders?

Phoenix earned the trust of AI engineers, ML engineers, and AI product teams because it was built around how they work: trace an AI application or agent, inspect its trajectory, run evaluations, investigate failures, curate datasets, compare experiments, and iterate. Teams that want those same workflows in a managed platform use Arize AX across development and production. Phoenix is open source and built on OpenTelemetry and OpenInference, the open-source instrumentation project and semantic conventions Arize created for AI observability. In June 2026, OpenTelemetry formally accepted a code grant of OpenInference GenAI instrumentation from Arize. That code is now being incorporated incrementally into OpenTelemetry’s GenAI instrumentation project, while OpenInference continues as an open, OpenTelemetry-compatible project.

Phoenix, OpenInference, and the communities around them are the foundation for what comes next. We plan to continue supporting the open, builder-first approach that has driven their adoption. Over time, we see an opportunity to connect the traces, evaluations, datasets, and experiments AI engineers already use in Phoenix and Arize AX with Dynatrace context across applications, infrastructure, user experiences, business processes, and outcomes. The goal is to preserve the workflows AI builders value while expanding the context available to them and strengthening the improvement loop across the AI lifecycle.

What comes next for Dynatrace and Arize?

With the acquisition complete, our teams can begin turning this shared vision into a product and platform roadmap. We’re pleased to welcome Arize CEO Jason Lopatecki, CPO Aparna Dhinakaran, and the entire Arize team to Dynatrace. As our teams come together, we will prioritize continuity for customers, developers, and the open-source community.

In parallel, our near-term product focus is identifying where connecting Arize’s tracing, evaluation, experimentation, and improvement workflows with Dynatrace context can create the most value across the AI lifecycle. We will shape our priorities with input from customers and builders and share updates as the work progresses.

If you’re an Arize customer, continue working with your account team for answers about what the acquisition means for you today. If you’re a Dynatrace customer exploring AI observability, start by mapping the AI applications and agentic systems your teams are building and scaling, and how you currently trace, evaluate, operate, and improve them.

If you’re building with Phoenix, keep building. The path forward is the one you’re already on.

Learn more about Arize.

Dynatrace, Arize, Arize Phoenix, the Arize logo, the Arize Phoenix logo, and the Dynatrace logo are trademarks of the Dynatrace, Inc. group of companies. All other trademarks are the property of their respective owners. Cautionary Language Concerning Forward-Looking Statements This blog includes certain “forward-looking statements” within the meaning of the Private Securities Litigation Reform Act of 1995, including statements regarding the expected benefits of the acquisition, capabilities expected to be available to organizations from using Dynatrace and Arize, and Dynatrace’s plans to integrate Arize into the Dynatrace platform. These forward-looking statements include all statements that are not historical facts and statements identified by words such as “will,” “expects,” “anticipates,” “intends,” “plans,” “believes,” “seeks,” “estimates,” and words of similar meaning. These forward-looking statements reflect our current views about our plans, intentions, expectations, strategies, and prospects, which are based on the information currently available to us and on assumptions we have made. Although we believe that our plans, intentions, expectations, strategies, and prospects as reflected in or suggested by those forward-looking statements are reasonable, we can give no assurance that the plans, intentions, expectations, or strategies will be attained or achieved. Actual results may differ materially from those described in the forward-looking statements and will be affected by a variety of risks and factors that are beyond our control, including our ability to successfully integrate Arize, the risks set forth under the caption “Risk Factors” in our most recent Annual Report on Form 10-K, subsequent Quarterly Reports on Form 10-Q, and our other SEC filings. We assume no obligation to update any forward-looking statements contained in this blog because of new information, future events, or otherwise.

Dynatrace, Arize, Arize Phoenix, the Arize logo, the Arize Phoenix logo, and the Dynatrace logo are trademarks of the Dynatrace, Inc. group of companies. All other trademarks are the property of their respective owners.

Cautionary Language Concerning Forward-Looking Statements

This blog includes certain “forward-looking statements” within the meaning of the Private Securities Litigation Reform Act of 1995, including statements regarding the expected benefits of the acquisition, capabilities expected to be available to organizations from using Dynatrace and Arize, and Dynatrace’s plans to integrate Arize into the Dynatrace platform. These forward-looking statements include all statements that are not historical facts and statements identified by words such as “will,” “expects,” “anticipates,” “intends,” “plans,” “believes,” “seeks,” “estimates,” and words of similar meaning. These forward-looking statements reflect our current views about our plans, intentions, expectations, strategies, and prospects, which are based on the information currently available to us and on assumptions we have made. Although we believe that our plans, intentions, expectations, strategies, and prospects as reflected in or suggested by those forward-looking statements are reasonable, we can give no assurance that the plans, intentions, expectations, or strategies will be attained or achieved. Actual results may differ materially from those described in the forward-looking statements and will be affected by a variety of risks and factors that are beyond our control, including our ability to successfully integrate Arize, the risks set forth under the caption “Risk Factors” in our most recent Annual Report on Form 10-K, subsequent Quarterly Reports on Form 10-Q, and our other SEC filings. We assume no obligation to update any forward-looking statements contained in this blog because of new information, future events, or otherwise.