Serval Raises $47M to Revolutionize IT Service Management with AI Agents

The Importance of Prestigious Clients for Startups

Many startups emphasize securing funding from well-known investors, however, attracting prominent customers is equally crucial for success.

This is a central point of emphasis for Serval, an enterprise AI company that recently announced a $47 million Series A funding round. Redpoint Ventures spearheaded the investment, with significant contributions from venture capital firms including First Round, General Catalyst, and Box Group.

Serval's Unique Approach to IT Service Management

Serval leverages agentic AI models to streamline IT service management, but distinguishes itself through a distinctive methodology. This approach harnesses the capabilities of agentic AI while mitigating potential drawbacks.

The company employs one agent dedicated to developing internal automations for routine tasks, such as software authorization and device provisioning. Founders envision this as a streamlined coding tool, supervised by an IT manager, yet capable of autonomous operation.

A separate agent handles user requests by activating these tools as needed, adhering to pre-defined rules established for each tool.

Simplifying Automation Tool Creation

According to Serval CEO Jake Stauch, a key objective was to simplify the process of creating automation tools.

“Our aim is to eliminate the perceived cost of building these automations,” Stauch explained to TechCrunch. “We want automating a process permanently to be simpler than performing it manually just once.”

Enhanced Security Through Agent Specialization

Dividing the process between a tool-building agent and a tool-using agent provides managers with enhanced control over permissions.

When an automation is created, managers define the conditions for its use, adding an additional layer of security against potentially overzealous help desk agents.

Enterprise clients are acutely aware of the risks associated with uncontrolled AI systems, influencing Serval’s decision to avoid a single, all-encompassing Help Desk Agent.

Preventing Rogue AI Actions

“The goal is to prevent scenarios where a user requests a harmful action, such as deleting all company data, and the AI agent immediately complies,” Stauch clarified to TechCrunch. “Instead, the agent should respond by stating it lacks the capability to perform that action, but can offer alternatives like password resets.”

The deterministic nature of the tools allows for the implementation of intricate permissions, such as requiring multi-factor authentication or restricting actions to specific timeframes.

Furthermore, an AI agent is readily available to modify the codebase and adjust these rules whenever necessary.

A Novel Approach to Agentic AI Oversight

This represents a new strategy for managing agentic AI systems. “Complete visibility and control over the AI agent’s actions are paramount,” Stauch states. “Serval facilitates this by enabling the creation of these tools and the customization of their associated permissions and approvals.”

Serval offers a robust solution for enterprises seeking to safely and effectively implement agentic AI in their IT service management processes.