Kantata debuts Agent Studio for building custom AI agents in plain language
SiliconANGLE Duncan Riley
Kantata launched Agent Studio so services teams can build AI agents by describing the job in plain English. It’s meant to cut the mess of one-off bots and keep client work consistent.
Based on reporting by SiliconANGLE, Duncan Riley — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Kantata has rolled out Agent Studio, a new way for professional services firms to build custom AI agents without writing code or filling out forms. The pitch is simple: say what you want done in plain language, and the system builds the agent inside Kantata’s Expertise Engine.
The timing says a lot about where this market is headed. Kantata argues that vendors are flooding services firms with narrow agents, each tied to one task like resourcing or risk reporting. That may sound efficient on paper. In practice, the company says, it creates a pile of little silos that all need to be maintained. Kantata is betting on a different model: one agent, called the Expertise Agent, that carries context across the platform.
Agent Studio works conversationally. A user describes the job, the Expertise Agent assembles the new agent and then refines it in the same back-and-forth flow. Before anything goes live, the system checks every build and edit for the right input and output structure. Once an agent is published, it can be triggered by a person or picked up automatically by a workflow with a single command.
The target audience is clear enough: practice leads and delivery directors who need recurring deliverables to look the same every time. Kantata says the point is to write a standard once, including format, escalation rules and outputs, and then apply it automatically instead of rebuilding it whenever staffing changes. That is the company’s answer to what it calls the “prompt tax,” the repeated work of re-explaining and reformatting the same task for an AI tool.
Chief executive Michael Speranza framed the product as an expertise play, not just an automation play. Chief technology officer Vikas Nehru took the governance angle even harder, saying an agent that improvises becomes a liability when the same deliverable has to stand up in front of a client every time. Kantata also points to Basware Oyj, which uses its software in its own professional services operation, and says Agent Studio is already available and in use. The company says more than 1,500 organizations use its software, after the 2021 merger of Mavenlink and Kimble Applications.
My take — AI-written commentary, not fact-checked reporting
This is the right instinct. Most AI agent talk still sounds like people trying to automate away the awkward part of professional services without noticing that the awkward part is the service. Kantata is at least admitting that expertise, not bot count, is the product. That’s a more honest pitch than “here’s another agent,” which is fast becoming tech’s version of a spare key under the mat.
Read more about this at: SiliconANGLE