Serval Launches Catalyst, an AI Agent That Automates Automation Building

Catalyst finds the repetitive work buried in a company’s ticket history and builds the automations to resolve it, because automating a task forever should be easier than doing it manually once.

Serval, the AI-native enterprise service management platform, launched Catalyst, the AI agent that automates how automations are built. Catalyst builds and configures automations across every area of the Serval platform, including workflows, skills, forms, access management, journeys, and more. The most notable of these is background agents: long-running agents that investigate, diagnose, and suggest fixes before an employee ever files a ticket.

“Most AI agents today wait for an employee to ask a question or submit a ticket. We believe the future is AI that acts before an employee ever submits a request,” said Jake Stauch, Co-founder and CEO of Serval. “With Catalyst, we’re automating automation itself – shifting enterprise service management from reactive support to proactive intelligence. By continuously improving your automation library, Catalyst reduces manual effort, enhances employee experiences, and frees teams to focus on higher-value work.”

Marketing Technology News: MarTech Interview with Mark Listes, CEO @ Pendulum Intelligence

Traditional automation platforms assume organizations already know what to build. Catalyst starts a step earlier, analyzing help desk ticket data to surface high-impact automation opportunities. Once gaps are detected, Catalyst drafts the workflows, skills, knowledge, and access configurations needed to automate the requests. All drafts are staged for administrators to review and test before publishing, so automation stays secure and governed.

Catalyst can also create background agents that discover user-impacting problems and propose fixes before anyone files a ticket. One customer discovered a network issue spanning two offices before anyone realized the incidents were connected. Looking across switch telemetry, DHCP data, and historical tickets, the agent correlated the incidents, ruled out hardware problems and wireless interference, and traced the issue to a network configuration drift. It then drafted a remediation workflow that admins could apply in one click.

Marketing Technology News: How MarTech Is Enabling Autonomous Brand Engagement Across Channels?

With Catalyst, non-technical functions like HR, Finance, and Legal can build automations on their own within an environment IT can govern. A few real-world examples:

• Automate manager and title changes: build workflows so managers can initiate a title change for a direct report or move them to a new team, with a message in Slack, Microsoft Teams, or email. The workflow automatically routes the request for the right approvals, then writes the update back to the HRIS, replacing a manual, multi-step process.
• Answer contract questions on demand: build workflows against the contract management system that answer employee questions about existing contract terms over Slack, Microsoft Teams, and email.
• Reclaim unused licenses: a background agent flags anyone provisioned for an app they haven’t opened in months, deprovisions the seat, and messages the person to re-request it if they still need it.
• Manage AI token spend: a background agent sends each manager a weekly view of their team’s spend across providers, while automating requests to increase usage limits.

Catalyst is built with enterprise-grade guardrails. It operates with the permissions of the person using it, so it can never see or touch anything that person couldn’t reach on their own, and it stays scoped to a single team workspace, aware only of that team’s data, automations, and integrations. Everything it builds is staged as a draft, so no automation touches employees or production systems until it’s explicitly published. Organizations decide who can publish automations and can require structured reviews with approval steps.

With this launch, Serval is redefining enterprise service management, transforming it from tracking requests to anticipating and preventing them. The result is a new approach to enterprise automation, where AI doesn’t just accelerate work; it turns every support function in an organization into a builder function.

Write in to psen@itechseries.com to learn more about our exclusive editorial packages and programs.

source https://martechseries.com/predictive-ai/ai-platforms-machine-learning/serval-launches-catalyst-an-ai-agent-that-automates-automation-building/