Cato Networks' use case reflects UiPath's agent vision

In the first round of agent AI applications, the most successful use cases were narrow and business-focused, as seen in Cato Networks Ltd.'s work with Zendesk.

Kato Networks is network security company It has its own Software-as-a-Service platform. We recently used agent AI to categorize tickets (specific requests made through the Zendesk IT portal) after they are closed.

Taqi Jaffri from UiPath and Daniela Rosenstein from Cato Networks discuss their collaboration.

“Tickets may be open for several hours to several days,” it said. Daniella Rosenstein (Photo, right), agent AI and RPA leader at Cato Networks. “We have many support engineers who review tickets throughout their lifecycle. Then we have one support engineer who closes the ticket and provides the appropriate classification. To make this data correct, we wanted to find the appropriate classification for every ticket.”

with Rosenstein Taki Jafri Senior Director of Product Management at UiPath Inc. (left) spoke to theCUBE Dave Bellante and rebecca knight in UiPath Fusionbroadcast exclusively on theCUBE, SiliconANGLE Media's live streaming studio. They discussed UiPath's vision for Cato Network's AI applications and autonomous agents. (*Disclosure below.)

Cato Networks develops task-specific agents

UiPath releases more agent AI updates Get access to a business automation platform that includes pre-built orchestration templates and Maestro Case Management for a variety of dedicated agents. One of UiPath's customer success stories is Cato Networks, which uses agent AI to streamline ticket classification.

“To get real ROI from agents,” Jafri said. “You have to have a very clear scope of what it does and an almost narrow and very specific definition of what you feed it or expect from it. That's how I would characterize it: small in scope, but large in impact.”

One of Cato Networks' agent use cases, which is still in development, involves analyzing tickets in the ecosystem and generating agents that can provide requested customer support. Using AI to address that part of the workflow saves employees time and mental energy, eliminating what Bellante calls “productivity paper cuts.”

“We found that about 60% of tickets were fairly repetitive,” Rosenstein explained. “People ask for licenses, and most of the time we have the same request, and we generated an agent, which we are currently testing… that can read tickets, review tickets, and ask for action.”

The rise of vibecoding has made it possible for non-developers to experiment with AI. Rosenstein foresees a world where executives can create agents within their company to take on specific tasks. Jaffri said UiPath allows developers to get creative with automation by building visual or video representations for agents. Techniques for explaining tasks to agents are currently the focus of innovation.

“These agents have new tools,” he said. “One of the things we announced at this conference is Project Delegate, where you can actually run a task and hit the screen record button, and your agent will watch it. You can categorize some tickets, and then you're like, “Here we go.''

The full video interview, part of SiliconANGLE and theCUBE's coverage, is below. UiPath Fusion:

(*Disclosure: TheCUBE is a paid media partner of UiPath Fusion. Neither UiPath Inc., the sponsor of theCUBE's event coverage, nor any other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE