Keenable, a startup founded by Andrey Styskin, former head of Yandex‘s search, AI, and cloud division, and AI scientist Matthias Petri, has emerged from stealth with $26 million in seed funding led by Accel, with participation from Conviction Partners and business angels. The company is building a web search index explicitly designed for AI agents and chatbots rather than human users, arguing that current search infrastructure is optimized for people who skim pages, while AI systems can process far larger portions of information and benefit from grounding responses in source documents.

Keenable says it has indexed more than 100 billion documents and its API is already in production at several AI labs and inference providers for both training and runtime use. The startup declined to name customers but recently partnered with voice AI company Gradium for live information retrieval. Styskin, drawing on 20 years of search experience at Yandex and Amazon, explains that the product differs from enterprise search solutions because those break down and become prohibitively costly at web scale. The key innovation is narrowing the search space quickly based on the query, rather than scanning the entire index. “If you do not fine-tune your index structures for a specific task, the cost of serving and scanning the whole internet is enormous because of the volume,” he told TechCrunch.

Accel partner Zhenya Loginov, who led the investment, notes that AI companies have few options for web-scale search infrastructure, especially as Google and Microsoft shut down existing search APIs to avoid cannibalization and move toward bundled, selective partnerships. Styskin saw early signals of this opportunity while at Amazon, where he worked with Petri on web search infrastructure for AI applications like Alexa. He observed Cloudflare data showing AI crawlers accounting for a growing share of search volume. His new startup, staffed with a handful of former colleagues, is also building proprietary retrieval capabilities, including an upcoming product called WebQueryLanguage that would help AI systems combine information from multiple web sources even when no single source contains the full answer.

Cost remains a major challenge. Styskin says building a giant search index is “painfully expensive,” but the startup is managing costs carefully. He acknowledges that while it is “extremely hard” to move people away from Google for traditional search, the innovators’ dilemma makes Google “beatable” on agentic queries, and a smaller company can innovate and offer more cost-efficient solutions to AI companies. Keenable has a team of 15 engineers across the U.S. and Europe and plans to double headcount by year-end to build its go-to-market motion.

The article places Keenable among other players like Brave and Exa, noting that Google itself is overhauling search for the AI era. Styskin’s conviction that the era of “ten blue links” is ending is shared even at Google, suggesting a broader industry shift toward search infrastructure built for AI agents rather than human eyeballs. The startup’s ambition is to become “the next Google for AI agents,” though many steps remain before that goal is realized.