Zetaris launches AI data harness with SpaceXAI tie-up

Zetaris has launched its cloud-based AI Data Harness and appointed Steve Nouri as an AI Strategy Advisor. It has also become a SpaceXAI Partner Network Transformation Partner.

The new product, Zetaris Cloud, is designed to let AI systems query enterprise data where it already sits across structured, unstructured and streaming sources, rather than requiring companies to move or duplicate that data first.

The system applies security, policy and access controls to data wherever it is queried. Zetaris says this is intended to address a common obstacle in enterprise AI projects: models and agents can generate code quickly but still struggle to access current internal data safely.

Vinay Samuel, Chief Executive Officer and Co-Founder of Zetaris, said the problem was slowing broader adoption of AI in business settings.

"Coding harnesses have dramatically accelerated software development, but enterprise AI still struggles to access trustworthy, governed data," Samuel said. "Our AI Data Harness gives AI tools a real-time path to the data they need, wherever it lives, without forcing companies to centralize everything first."

Zetaris Cloud offers the architecture as a cloud service, allowing customers to connect models and agents to existing data environments without committing to a single cloud provider, data platform or AI model.

Zetaris also argued that reducing the need to copy and centralise data could lower infrastructure demand. Based on an internal customer assessment, it said the product could cut total cost of ownership by up to 67 per cent.

Partnership and testing

Alongside the product launch, Zetaris outlined a broader collaboration with SpaceXAI centred on a large online developer contest run by GenAI Works. More than 5,000 developers from over 100 countries are expected to take part, using SpaceXAI's Grok Build with the Zetaris AI Data Harness.

The event is framed as a test of whether AI coding tools can be paired with direct access to distributed enterprise data in live development settings. Developers will build enterprise AI applications using both systems together.

"Putting these technologies in the hands of thousands of developers will show what becomes possible when AI agents can work with live enterprise data without sacrificing security," Samuel said. "The hackathon is designed to move that idea from theory into working applications."

SpaceXAI described the pairing as a combination of coding tools and governed data access.

"A coding harness is an important part of building enterprise AI applications. Also very important is instant access to relevant, governed data-the role of a data harness. Through the hackathon, developers will explore how Grok Build and Zetaris can be used together to combine these two complementary harnesses to build AI-enabled workflows. We look forward to seeing what the developer community creates," Adams said.

Nouri joins

Nouri joins Zetaris while remaining Chief Executive Officer and Co-Founder of GenAI Works, which organises the Open Agent Hackathon. According to the company, he has built a large public profile in the AI sector and has more than two million followers across his platforms.

His previous roles include leading data science teams at CSIRO's Data61 and the Australian Computer Society. He is also a member of the Harvard Business Review advisory council and founder of AI4Diversity, a non-profit focused on underrepresented talent in AI.

Zetaris said Nouri's appointment reflects its push to bring in higher-profile figures as it expands beyond traditional enterprise software circles.

"AI is moving fast, but the data infrastructure underneath it hasn't kept pace, and that gap is exactly where the real work needs to happen," Nouri said. "What drew me to Zetaris is that the team isn't just talking about that problem, they've built a way to solve it. Hosting the Open Agent Hackathon through GenAI Works and bringing that community together with Zetaris and SpaceXAI is a particularly compelling moment to get involved directly."

Data bottleneck

Zetaris is positioning the product around a familiar problem for companies trying to deploy AI beyond pilot projects. While software development tools can now generate working code quickly, corporate data often remains spread across cloud, on-premises and edge systems, with differing formats and governance rules.

That fragmentation can make it difficult for AI agents to work with accurate, current information, especially where businesses do not want to centralise data into a single repository first. Zetaris said its system is designed to sit across those environments and allow queries in place.

Samuel also linked the issue to infrastructure costs tied to repeated data duplication across systems.

"Every time enterprises copy and centralize data to build something new, they duplicate storage, compute, and networking," Samuel said. "The world holds roughly six times more data than it actually needs, and most of that is duplication chasing a premise that was already flawed. AI is going to make that duplication far more expensive unless the underlying model changes."