QueryStory wants you to believe what AI is telling you

Shapor Naghibzadeh says the idea for QueryStory comes from a very old problem: figuring out what’s true when the data is messy and the stakes are high. He learned that lesson during the Operation Aurora cyberattacks at Google in 2009, when he was pulled into a war room to explain what was happening across the company’s servers. Later, after years building tools for security analysts, he started thinking the same approach could help people make sense of enterprise data with large language models.

That’s the bet behind QueryStory, the startup he co-founded with Stanley Yang, a former Google colleague and lead engineer at EvolutionIQ, and David Glusic, who came from Accenture. The company came out of stealth today after raising a $6 million seed round in late 2025 from Brightmind Ventures and New York Life Ventures at a $60 million valuation. Since then, it has been piloting its product with customers.

The target customer is not a hobbyist or a lone analyst. QueryStory is aimed at large enterprises with big proprietary databases, especially teams like sales and operations that need to ask questions, review answers, and keep a record of how those answers were reached. New York Life Ventures partner Tim Del Bello said he’s using it to replace work that used to take several people and hopes to turn a quarterly business review into a real-time dashboard.

What QueryStory is really selling is not just answers, but proof. In a test with a space-activity database, it produced visualizations and analysis in a few hours, where a similar project once took weeks with a developer. It also surfaced a confidence indicator explaining why the system believed its own analysis, along with the SQL queries behind it. That matters because the frontier labs’ own coworking tools are useful but intentionally limited, and enterprises keep running into the same problem: once AI starts answering questions for hundreds or thousands of employees, the company gets a pile of slide-deck truth with no easy way to trace it back.

Naghibzadeh argues that purpose-built software can do better than a generic chat window because it preserves context and keeps humans in the loop when needed. QueryStory is model-agnostic, though it mostly uses the latest frontier models for now, and its pitch is blunt: don’t buy more AI noise than the business can trust.