Keenable, a startup building independent web search infrastructure for AI, has come out of stealth with $26M in seed funding. Accel led the round, with Conviction Partners and angels from Amazon, ClickHouse, Databricks, Google, Snowflake and SpaceXAI also participating.
The company was founded in 2025 by Andrey Styskin, who previously ran Yandex’s search, AI and cloud division, and German AI scientist Matthias Petri. The pair argue that search engines were built for people who cannot scan whole pages, while AI systems need different retrieval primitives entirely. Keenable says it has assembled a web index of more than 100 billion documents, and its API is already in production at several AI labs and inference providers during both training and runtime.
The startup recently struck a partnership with voice AI company Gradium for live information retrieval, and it is developing Web Query Language, a system meant to let AI combine answers from multiple web sources when no single page holds the full answer. Styskin, drawing on two decades of search work at Yandex and Amazon, says the goal is to make web-scale search an order of magnitude more cost-efficient than traditional infrastructure.
The timing helps: Google and Microsoft have been tightening access to their search APIs to avoid cannibalization, leaving AI companies with few options for grounding models in the live web. With 15 engineers across the US and Europe, Keenable plans to roughly double headcount by year end to build out its go-to-market motion.
