Startup Keenable has announced its emergence from stealth after raising $26 million in a seed funding round led by Accel, with participation from Conviction Partners and several angel investors. The company is building web search and indexing infrastructure for AI systems, rather than search engines designed primarily to meet the needs of human users.
Keenable is led by Andrey Styskin, who previously headed search, AI and cloud divisions at Russian company Yandex, and German AI scientist Matthias Petri. The company’s team currently has 15 employees distributed across the United States and Europe, and it plans to use the new funding to double its headcount by the end of the year, in addition to developing its commercial operations and customer outreach.
An index exceeding 100 billion documents
Keenable says it has built a web search index containing more than 100 billion documents, and that its API is already being used in production environments by several AI labs and inference service providers, during both training and live operation. The company has not disclosed the names of its customers, but it recently announced a partnership with AI-powered audio company Gradium to support direct information retrieval.
The company believes conversational systems produce better results when their answers are grounded in source documents. This use differs from the model of traditional search engines, which were designed to help humans quickly access a limited number of pages, whereas intelligent models can read and process larger portions of content to carry out more complex requests.
What is the company trying to change?
According to Styskin, the challenge lies not only in collecting web content, but also in rapidly narrowing the search space in a way suited to the type of query. Scanning the entire internet or running an index not tailored to a specific task can be extremely costly at this scale. That is why Keenable is developing its own indexing structures and retrieval capabilities, as well as an upcoming product called Web Query Language, which aims to enable AI systems to assemble an answer from multiple web sources when no single document contains the complete answer.
Why does this news matter?
Keenable’s funding round comes at a time when AI companies need broad and continuous access to the web, while ready-made search options are becoming less clear. According to Accel partner Zhenya Loginov, Google and Microsoft are taking steps to shut down or restrict existing search APIs to avoid competing with their core services, and are instead moving toward more integrated models and selective partnerships.
This creates an opening for independent infrastructure providers, but it does not guarantee their success. The company faces high costs to build and operate a massive index, and the market includes competitors such as Brave and Exa, while Google itself is redesigning the search experience for the AI era. Therefore, the confirmed information at present is that Keenable has secured funding and built infrastructure used in production; its ability to provide cheaper search or become “Google for AI agents” remains a future goal not established by the available evidence.