Israeli Startup Keewano Launches KeewanoDB for AI Agents with $12 Million Seed
Israeli startup Keewano launched KeewanoDB for machine reasoning, securing $12 million in a seed round led by Hetz Ventures to support AI agent architectures.
The Israeli startup Keewano has announced the launch of KeewanoDB, the first database built from the ground up for large-scale machine reasoning.
Along with the product launch, the company revealed a $12 million seed funding round led by Hetz Ventures, with participation from a16z speedrun, Remagine Ventures, DIG Ventures, and private investors.
The technology developed by the company aims to solve a major bottleneck in the world of artificial intelligence. While traditional databases were designed for humans and queries on flat tables, AI agents require a complete sequence of events in real-time and with broad context.
KeewanoDB allows processing trillions of events in parallel and providing AI agents with answers based on precise context, while eliminating event-based pricing that has been common in the industry until now.
Keewano was founded in 2024 by Mark Kardashov (CEO), Dima Karger, Pavel Bibergal, and Vitaly Bukhovsky. The four founders bring rich industry experience, with Kardashov and Bukhovsky sharing two previous joint exits: selling TestProject to Tricentis in 2019 and selling Devalore to Abra in 2022. Additionally, Bibergal served as CTO at the gaming company Plarium, and Karger managed one of its leading studios.
Transforming Data Architecture for AI Agents
When striving to use AI agents to answer complex business questions, you discover that existing databases were designed for humans and not for machines that need to draw conclusions in real-time, said Mark Kardashov, co-founder and CEO of the company.
In 1964, Japan did not try to speed up trains on the same track, but built a completely new track. AI agents need a completely different data architecture, and that is exactly what we built.
Yehuda Taub, managing partner at Hetz Ventures, added that every wave of AI adoption exposes the underlying layer that was not built for the new workloads, and that Keewano addresses this gap from the root.
