Zetaris launches a data harness for AI agents to query enterprise systems in place

Zetaris launched Zetaris Cloud and its AI Data Harness on October 6, a product the company says lets AI agents query enterprise data in place across cloud, on-premises and edge systems via SQL, programming interfaces and MCP. Zetaris claims the harness can cut total cost of ownership by up to 67 percent, a figure the launch release attributes to an internal assessment of one enterprise customer and not an independently verified benchmark. Co-founder and CEO Vinay Samuel, who founded Zetaris in 2013, said the product gives AI tools a real-time path to data wherever it resides without requiring companies to centralize everything first.

Zetaris launches a data harness for AI agents to query enterprise systems in place Co-founder Vinay Samuel argues that governed access to scattered company data will determine whether enterprise AI is dependable. By Ryan Merket https://runtimewire.com/author/ryan-merket · Published Primary source: PR Newswire https://www.prnewswire.com/news-releases/zetaris-launches-cloud-based-ai-data-harness-giving-ai-agents-a-safe-real-time-path-to-enterprise-data-302899905.html Why it matters Enterprise agents need access to governed, current data to work beyond demos. Zetaris is betting that querying data in place can reduce the integration work and duplication that stand between prototypes and production. Zetaris launched Zetaris Cloud and its AI Data Harness https://www.prnewswire.com/news-releases/zetaris-launches-cloud-based-ai-data-harness-giving-ai-agents-a-safe-real-time-path-to-enterprise-data-302899905.html?ref=runtimewire on October 6th, pitching them as a way for AI agents to query enterprise data where it already lives instead of waiting for it to be copied into a central system. Co-founder and CEO Vinay Samuel https://www.zetaris.com/zetarisfounders?ref=runtimewire , who founded Zetaris in 2013, has spent his career on the problem behind that pitch: making analytics work across complicated data environments. Samuel's earlier work spanned event-based analytics and parallel-database technology. Zetaris says he held leadership roles connected to Netezza, Greenplum, Vertica and Gridmatics in Asia before starting the company with Michael Cimino. The company began with an enterprise database and analytics platform. Zetaris Cloud and the AI Data Harness apply that approach to AI agents. The data access problem Zetaris is selling against AI coding tools can generate software quickly, but enterprise applications still need accurate, current information from systems that often remain separate. Zetaris says its harness queries structured, unstructured and streaming data across cloud, on-premises and edge systems without requiring a preliminary centralization project. It says security policies and access controls apply as agents query that data. The architecture is a bet on federation: leave data in place and route queries to it, rather than first building and maintaining another consolidated copy. Zetaris Cloud is the cloud-delivered version of that approach. Zetaris' product site https://www.zetaris.com/cloud?ref=runtimewire says the service runs in the customer's cloud account, with customers retaining control of the cluster, keys and infrastructure bill. The platform describes access through SQL and programming interfaces as well as MCP. A working agent demo can be built against a prepared sample; deploying an agent against live company information raises harder questions about permissions, data freshness and inconsistent definitions across systems. Zetaris is selling the layer that it says can answer those questions without first moving all the underlying data. The launch release https://www.prnewswire.com/news-releases/zetaris-launches-cloud-based-ai-data-harness-giving-ai-agents-a-safe-real-time-path-to-enterprise-data-302899905.html?ref=runtimewire says the product can reduce total cost of ownership by up to 67 percent. The footnote attributes that figure to an internal assessment comparing costs before and after an enterprise customer adopted the product. It is a company-reported result, not an independently verified benchmark, and the announcement does not describe the customer sample or assessment method. The figure does not establish what customers generally should expect. Zetaris adapts its data-platform thesis for AI agents Samuel described the product as giving AI tools a real-time path to data wherever it resides, without requiring companies to centralize everything first. That emphasis fits Zetaris' earlier data-platform work: the company says it was built to simplify analytics and let organizations use data without moving it first. The AI launch reframes that existing approach for a new user of enterprise data: the agent, which needs enough context to act but still has to operate within company controls. Zetaris is also tying the launch to a developer event. The company says it will headline the Open Agent Hackathon with SpaceXAI, pairing Zetaris' data harness with SpaceXAI's Grok Build. Zetaris' release describes a 144-hour event from October 22nd to 27th and says more than 5,000 builders from over 100 countries are expected. The organizer's event page https://academy.genai.works/courses/open-agent-hackathon-2026/details?ref=runtimewire , however, lists the build window as October 15th to 20th. The dates conflict; the organizer's schedule places the event earlier than the company's release does. The event is intended to put the integration in front of developers building agents that use distributed data. Participation figures are forecasts, not a record of completed attendance, and a hackathon can show how quickly teams connect tools without establishing that those connections will meet enterprise security and reliability requirements in production. For Zetaris, it is a way to turn an architectural argument into developer experiments while the company launches its cloud service. The announcement also names Steve Nouri https://www.linkedin.com/in/stevenouri/?ref=runtimewire , CEO and co-founder of GenAI Works, as Zetaris' new AI Strategy Advisor. Nouri has built an AI education and community business and founded AI4Diversity; the University of Technology Sydney https://www.uts.edu.au/for-students/experience-uts/alumni/steve-nouri?ref=runtimewire says he completed a master's degree in information technology in 2018 and previously worked in AI and data science roles at the Australian Computer Society and CSIRO's Data61. His appointment connects Zetaris' product launch to GenAI Works' developer community and the hackathon it organizes. Zetaris also says it has joined the SpaceXAI Partner Network as a Transformation Partner. Those partnership and event details are claims in Zetaris' announcement. The test for Samuel's bet will be whether developers can use the harness to build agents that reach live data under real access policies, and whether enterprises find that approach simpler and less costly than building centralized data pipelines. The announcement provides no customer deployment results or independent cost measurements to assess the case.