Serval raises $47M to bring AI agents to IT service management

Some startups pride themselves on having prestigious financial backers — but having prestigious customers is just as important.

That’s a major point of pride for Serval, an enterprise AI company that announced a $47 million Series A round on Tuesday. The round was led by Redpoint Ventures, with participation from heavy-hitting venture firms such as First Round, General Catalyst and Box Group. But even more impressive than the funders is the company’s client list, which includes big AI players like Perplexity, Mercor and Together AI.

In broad strokes, Serval is using agentic AI models to automate IT service management, but the company has a unique approach that takes advantage of the power of agentic AI and avoids many of its pitfalls. An agent is used to code internal automation for everyday tasks, such as approving software or provisioning a device. The founders see it as a kind of vibe-coding tool, overseen by an IT manager, but doing most of the work themselves. An individual help desk agent calls those tools commands, responding to user requests by following the rules established by the tools.

Serval CEO Jack Stauch says the key was to make the process of creating a tool as simple as possible.

“We don’t want them to experience the marginal cost of building these automations,” Stauch told TechCrunch. “We want to make it easier to automate something forever than to do it manually once.”

Splitting the task into two agents—one to create the tools and one to use them—also gives managers a way to keep track of permissions. When an automation is created, the manager will set rules for when it can be used, which provides an additional line of defense against redundant help desk agents.

Enterprise clients are acutely aware of the risks of a rogue AI system, which is part of why Serval decided against a single all-purpose help desk agent.

“You don’t want someone to go to Slack and say, ‘Hey, I want to delete all the company data,’ and the very helpful AI agent replies, ‘Great, I’ll delete all the data,'” Stach told TechCrunch. “Instead it will say, ‘Hey, I don’t have a tool to delete all the company data. But I do have a tool to reset your password or do one of these other things.'”

Because the tools themselves are deterministic, they can include very complex permissions, such as allowing certain actions after a specific multi-factor authentication process or within a specific time period. And anytime these rules need to be changed, there’s an AI agent ready to dive into the codebase and change it

This is a new approach to the very general problem of how to supervise agentic AI systems. “You want full visibility and control over what the AI ​​agent is doing,” says Stach. “And you do that by using Serval to build those tools and customize the permissions and authorizations behind them.”