Keenable, a startup co-founded by Andrey Styskin — who previously led search, AI and cloud at Yandex — and German AI scientist Matthias Petri, has emerged from stealth with $26 million in seed funding. The round was led by Accel, with participation from Conviction Partners and several business angels.
The problem they address
Traditional search engines were designed for human users who scan pages for short answers. As AI chatbots and agents increasingly perform web searches and tasks, they can consume and reason over much larger portions of content. Keenable argues this shift requires web infrastructure and indexes built specifically for AI-driven retrieval rather than human-oriented ranking alone.
What Keenable is building
The company says it has constructed a web search index containing more than 100 billion documents. Its API is already used in production by several AI labs and inference providers for both training and runtime use. Keenable declined to name customers, but recently announced a partnership with voice AI company Gradium to support live information retrieval.
Styskin says AI chatbots perform better when their answers can be grounded in source documents, creating a different kind of growth loop than traditional search driven by human behavior. Keenable is developing proprietary retrieval capabilities and an upcoming product called WebQueryLanguage, intended to enable AI systems to combine information across multiple web sources to answer questions even when no single source contains the full answer.
How this differs from enterprise search
Drawing on 20 years building search at Yandex and Amazon, Styskin explained that enterprise search solutions can become prohibitively expensive at web scale unless index structures are fine-tuned for a specific task. Serving and scanning the entire internet carries enormous cost due to volume; Keenable focuses on narrowing the search space very quickly based on the query to reduce serving and scanning expenses.
Market context and competition
Accel partner Zhenya Loginov, who led the investment, noted that AI companies have limited options for web-scale search infrastructure, especially as Google and Microsoft move to restrict existing search APIs to avoid cannibalization. The big tech firms are favoring more bundled approaches and are selective about partnerships.
Styskin said these industry moves confirmed an opportunity he first saw while at Amazon. He cited Cloudflare data indicating AI crawlers are responsible for an increasing share of search volume, reinforcing the case for web search infrastructure built with AI in mind. Other entrants in the space include Brave and Exa, and Google itself is reworking its search experience for the AI era.
Costs and team plans
Building a large search index is costly: as Styskin put it, "don’t ask — it is painfully expensive." Keenable is managing costs deliberately and pacing growth. The company currently has 15 engineering staff across the U.S. and Europe and plans to use the new capital to double headcount by the end of the year and develop its go-to-market motion.
Outlook
Keenable aims to become a primary web-search infrastructure provider for AI agents, but faces several hurdles before it could match major incumbents on scale and adoption. The market is evolving quickly as legacy search providers change product strategies and new competitors appear, yet the trend is clear: search infrastructure is adapting to the needs of agents as well as humans, and Keenable is positioning itself to serve that shift.
