Observability in AI: How Dynatrace adapts to agents – News

Dynatrace Inc. built its place in the technology world by providing observability and security for traditional workloads. That mission has now come to include support for artificial intelligence workloads with AI-powered observability, which has given the company a prime position from which to observe the latest trends.

“We see a change in how people are building applications,” said Alois Reitbauer (pictured), chief technology strategist of Dynatrace. “In the past it was basically OpenAI, you used OpenAI and then it started to switch to other models. Now, we see people experimenting way more, like A-B testing models and the practice of … AI native engineering.”

Reitbauer spoke with theCUBE’s Rob Strechay at the KubeCon + CloudNativeCon NA event, during an exclusive broadcast on theCUBE, News Media’s livestreaming studio. They discussed how AI adoption is transforming enterprise observability. (* Disclosure below.)

Observability adjustments for agentic AI

Rapid adoption of AI models in the enterprise world has led to an adjustment by observability firms such as Dynatrace. While the models themselves may be improving in transparency, the advent of agentic AI is adding a whole new layer of complexity in the observability world, according to Reitbauer.

“One of the very positive trends was that AI models now actually share a train of thought that helps you to understand how the model arrived at where they arrived,” he explained. “That helps us a lot on the debugging side. But debugging AI in agentic applications is kind of different. The more we move into more dynamic systems, like going more into this agentic world, the more the individual transactions will be different.”

Dynatrace announced its next-generation cloud operations solution for Microsoft Azure that included support for the platform’s agentic AI initiatives. Dynatrace is focused on bringing observability to agentic operations, Reitbauer emphasized, because the task-oriented nature of the autonomous software does not always allow for traditional monitoring.

“Guardrails are a key and guardrails started to emerge very early on,” he said. “Really thinking to the next step about agentic, we have to track against goals and I think that’s where business observability comes in. You’re delegating a task, you’re not looking AI over the shoulder.”

Here’s the complete video interview, part of News’s and theCUBE’s coverage of the KubeCon + CloudNativeCon NA event:

(* Disclosure: Dynatrace sponsored this segment of theCUBE. Neither Dynatrace nor other sponsors have editorial control over content on theCUBE or News.)

Photo: News