#Israel #Tel Aviv #AI Agents #KeewanoDB #Machine Reasoning

KeewanoDB: The Next Generation in Machine Reasoning

In a remarkable step for database architecture, Keewano has officially launched KeewanoDB, a pioneering database specifically designed for large-scale machine reasoning. Supported by a $12 million funding round led by Hetz Ventures, alongside a16z speedrun, Remagine Ventures, DIG Ventures, and other angel investors, KeewanoDB is set to transform the way AI agents operate.

The Problem with Traditional Databases

For years, analytical systems have catered primarily to human users, focusing on storing data in flat formats such as rows and columns. This legacy limits how organizations can analyze data because they frequently have to reconstruct the context during query time. This not only slows performance but also results in higher operational expenses as the complexity increases.

Most existing databases manage only a limited number of event types due to the extra instrumentation required and the subsequent slowing of queries. The result is that critical context gets discarded in favor of data retention—often hindering the understanding of data flows.

Mark Kardashov, Keewano's Co-founder and CEO, stated, "Our traditional databases weren't engineered for the kind of complex, context-driven questions that AI needs to answer. Instead of simply querying tables, we need databases that provide real-time, complete context to support AI-driven reasoning."

KeewanoDB's Unique Architecture

KeewanoDB distinguishes itself by adopting a radically different architectural framework. Instead of separating events into isolated rows and columns, it retains every entity's complete event sequence in order. This revolutionary approach enables AI agents to directly access raw data, transforming queries into much faster and accurate outcomes.

Gone are the days when AI agents could only deliver high-level statistics. With KeewanoDB, agents can probe deeply into the historical sequences of user actions, deciphering why a customer might have left or what patterns exist across user behaviors. For example, it can illuminate pathways that led to outcomes never anticipated by human analysts—exceptionally useful for improving user retention and enhancing customer experiences.

With the ability to parse through 250 million events in under half a second, KeewanoDB furnishes results tailored for immediate reasoning by AI agents. An agent's ability to query vast data histories enables organizations to act efficiently and adapt to emerging trends proactively.

A New Pricing Model for Scalability

Moreover, KeewanoDB introduces a pricing model devoid of per-event costs, allowing organizations to capture and analyze more data without worrying about escalating expenses. This model significantly simplifies the integration process and can either complement existing data warehouses or serve as a replacement.

Users have the option to deploy their custom AI agents or leverage Keewano’s provided solutions. The architecture is designed to adapt how databases communicate on a fundamental level, aligning with the growing needs of modern AI applications.

Founders with Proven Expertise

Keewano was co-founded by a seasoned team comprising Kardashov, Dima Karger, Pavel Bibergal, and Vitaly Bukhovsky, who collectively have decades of experience in analytics and software engineering. Their background spans successful ventures, including two previous exits and extensive work in gaming analytics at scale. This expertise uniquely positions Keewano to address the needs of AI development and machine reasoning.

A Vision for Future AI Implementations

Judah Taub, Managing Partner at Hetz Ventures, articulated the significance of this innovation by saying, "The emergence of machine reasoning underscores a gap in analytics infrastructure. KeewanoDB targets this gap directly, offering a next-generation solution rather than merely patching existing systems."

Kardashov further elaborated, "Just as Japan constructed a new rail line rather than trying to improve existing tracks in 1964, we recognize that AI agents necessitate a radically different database architecture. KeewanoDB is that new line, built precisely for tomorrow's demands."

In conclusion, KeewanoDB represents not just a remarkable achievement in database technology but also a substantial leap towards a future where AI and machine reasoning can work together seamlessly and intelligently. Organizations aiming to leverage AI at scale now have an innovative framework that eliminates traditional barriers, pushing the boundaries of what is analytically achievable.