NEW YORK, New York — October 7, 2025 — Leads & Copy —Dataiku has launched Agent Hub, a centralized workspace that allows employees to discover, use, and create approved AI agents, while IT maintains control over model and data access, as well as the agent lifecycle. The new platform aims to eliminate fragmentation from isolated agent experiments, helping enterprises unlock agent business value and determine ROI through visibility into adoption, usage, and impact.
Enterprises are adopting AI agents, but often encounter issues like scattered experiments, limited IT oversight, and difficulties in proving compliance and ROI. These challenges can undermine trust and hinder innovation.
To address these problems, Agent Hub provides a collaborative workspace within Dataiku’s Universal AI Platform. It enables employees to build and use AI agents, with IT maintaining oversight of access, governance, and lifecycle management. Agent Hub securely connects AI agents to business data and models, orchestrates them across teams, and governs them end-to-end.
Agent Hub offers several benefits:
-
IT Control: IT can manage agent, model, and tool availability to maintain compliance.
-
Trusted Library: Employees can find and use approved agents in a central location.
-
Fast Building: Users can quickly create and share agents using a simple builder and templates.
-
Scalability: IT can identify top-performing agents and promote them across the company.
According to Dataiku co-founder and CEO Florian Douetteau, Agent Hub balances employee freedom with IT control, turning fragmented projects into demonstrable ROI.
Agent Hub integrates with existing company data and tools, enterprise data pipelines, and ML workflows, with built-in security and oversight. Dataiku co-founder and CTO Clément Stenac stated that Agent Hub bridges the gap between experimentation and enterprise-grade adoption, allowing companies to generate value that transforms their business.
Agent Hub is available to all customers on the latest release of The Universal AI Platform. For more information or a demo, visit www.dataiku.com.
Contact: Gina King, press@dataiku.com
Source: Dataiku
