Swivel is a prime example of a vertical AI agent application within the AdTech stack. It moves beyond simple automation by implementing what it calls an "agentic operating layer," where AI agents interpret operator intent to execute complex tasks across multiple platforms. This puts Swivel firmly in the application layer of the agent ecosystem, specifically serving the publishing and digital media sector.
They matter to the ecosystem because they demonstrate how agents can be used to bridge "legacy" digital systems that lack modern, unified APIs. By acting as the glue between ad servers, CRMs, and performance tools, Swivel shows that the most immediate value for AI agents often lies in solving the "swivel-chair" problem—automating the manual data movement and decision-making that currently bogs down enterprise workflows.
Swivel is a New York-based technology company that addresses the operational friction inherent in digital advertising. Founded by Joe Hirsch, Matt Dearborn, and Rich Lin, the company rebranded in 2024 from its previous identity, PilotDesk, to signal a broader focus on AI-driven orchestration. The name Swivel is a direct reference to "swivel-chair execution," a term used in ad operations to describe the manual, error-prone process of copying data between disparate systems like Google Ad Manager, CRM platforms, and various supply-side interfaces.
Modern digital media companies manage a mix of direct sales, programmatic deals, and private marketplace (PMP) agreements. Each of these requires specific setup procedures, performance monitoring, and constant adjustments to optimize yield. For many publishers, this work remains manual, consuming hundreds of hours of human labor for tasks that are logically consistent but technically fragmented. Swivel provides a no-code layer that sits on top of these systems, allowing teams to translate their monetization strategies into automated actions.
The platform operates as an agentic layer that drives decisions and actions across the ad tech stack. Unlike traditional automation, which often relies on brittle, fixed scripts, Swivel uses an engine that allows operators to set specific parameters and guardrails. Within these boundaries, the AI agents handle line item creation—the process of defining which ads appear to whom—and trafficking, which involves the actual technical setup and launch of ads in the server.
By processing live performance data in real time, the system can inform monetization decisions without requiring a human to manually pull reports and re-input data. This continuous action loop is designed to reduce the lag time between identifying a performance issue and executing a fix. For publishers, the efficiency gain is measured in both headcount reduction and revenue growth, as the agents can pull levers faster than manual operators to capture shifting demand.
In April 2024, Swivel announced a $5.8 million Series A funding round led by Tribeca Venture Partners and Ardent Venture Partners. This capital infusion was intended to accelerate the development of deeper AI-driven orchestration features. The company occupies a specific niche within the agent ecosystem: vertical application of agents to a high-complexity, high-volume data environment. While general-purpose AI agents struggle with the specific jargon and API nuances of ad tech, Swivel is built to navigate the particular complexities of ad delivery.
Led by CEO Joe Hirsch, the team brings a background in media and ad tech that is reflected in the product’s focus on the actual day-to-day pain points of publishers. The 16-person team operates out of New York, a strategic location given the city's status as a hub for both the media and advertising industries. As publishers face increasing pressure to optimize revenue while containing costs, Swivel’s pitch is that technology should handle the rote execution, leaving strategy to the human operators.
An agentic AI operating layer that automates manual ad operations and monetization workflows.