That difference matters because an agent can generate far more search activity than a typical human session. A system working through a complex request may gather information, analyze what it finds and issue additional searches based on those results. Keenable argues that relying on conventional search and page-retrieval APIs for that workflow can quickly become expensive when applications operate continuously at scale.
The company is also positioning its independent index as an alternative at a time when major technology providers, including Google and Microsoft, have restricted access to some third-party search APIs. Keenable believes AI developers will need infrastructure that allows their systems to retrieve fresh web information without depending entirely on search products originally designed for human use.
According to Keenable, its architecture can make AI-driven searches roughly an order of magnitude more cost-efficient than traditional search infrastructure. Lower retrieval costs could allow agents to perform more searches while working through complicated tasks without making latency or infrastructure spending prohibitive.
Keenable is building additional retrieval technology on top of its search index. One of its first products is Web Query Language for AI, a system designed to help models collect and combine information from thousands of web sources before reasoning over the results.
The product is aimed at questions where the necessary information is distributed across multiple webpages rather than contained in a single document or database. Instead of returning isolated pages, Web Query Language is intended to let an AI system assemble relevant material from several sources and work across the combined information. Keenable plans to introduce Web Query Language publicly along with several additional products over the coming week.
The company was founded in 2025 by Andrey Styskin and Matthias Petri, both of whom previously worked on large-scale search systems. Styskin previously served as CEO of Yandex Search before becoming a director at Amazon AGI, while Petri was a Principal Applied Scientist at Amazon AGI.
Keenable is the founders’ third effort involving a web-scale search index. At Yandex, Styskin helped develop an index containing about 200 billion documents. He and Petri later worked on web search infrastructure at Amazon AGI that was used internally across Amazon.
Keenable says its API is already being used in production by several AI labs and inference providers. The company is working with model developers on live information retrieval during both model training and runtime, although it has not disclosed the customers involved.
The broader technical bet is that stronger access to external information can improve what AI systems are able to do without requiring every relevant fact to be stored inside a model’s parameters. Keenable wants models and agents to retrieve more current and useful information as part of the reasoning process itself.
“Today’s leading AI models are excellent summarisation machines. They can answer almost any question, but ask them about a subject you know deeply, and you’ll quickly find their limits. The next frontier, that will take average answers to outstanding ones, will come from giving AI dramatically better access to the world’s knowledge. Our goal is to make it accessible at AI scale.”
The new capital will be used to expand Keenable’s engineering organization across the Bay Area and Europe. The company also plans to invest in its crawling systems, enlarge its independent web index and continue developing the retrieval technology behind Web Query Language.
This analysis is based on reporting from Pulse 2.0 & TechCrunch.
Image courtesy of Keenable.
This article was generated with AI assistance and reviewed for accuracy and quality.
