A startup wants to rebuild web search from the ground up, this time for machines instead of people. Keenable just came out of stealth with $26 million in seed funding, and according to TechCrunch AI, the round was led by Accel with participation from Conviction Partners and a group of angels. The pitch is simple to state and brutally hard to execute: the internet was indexed for humans who skim, but AI agents read and process far more, so the infrastructure underneath needs a rethink.
The founders bring real search pedigree. Andrey Styskin previously ran Russian search giant Yandex’s search, AI, and cloud division, and he spent years on search infrastructure at Amazon, including work tied to Alexa. His co-founder, German AI scientist Matthias Petri, worked alongside him on that AI-focused retrieval problem. TechCrunch AI reports the company has already built an index of more than 100 billion documents, with an API running in production at several AI labs and inference providers during both training and runtime.
Why this matters
What stands out here is the shift in who search is actually serving. Styskin’s argument is that chatbots perform much better when they can ground their answers in source documents, which he says creates “a new flywheel that is different from what Google learned from human behavior.” Google spent two decades optimizing for click patterns and human attention. Agents don’t behave like that. They pull large chunks of context, cross-reference, and synthesize. Search tuned for a person scanning ten blue links isn’t the right tool for that job.
The timing is deliberate. Accel partner Zhenya Loginov, who led the deal, points out that AI companies have very few options for web-scale search infrastructure, especially as Google and Microsoft close down their search APIs to avoid cannibalizing their own products. The giants are moving toward bundled deals and hand-picked partners. That leaves a gap, and Keenable is running straight at it.
The technical challenge
Scanning the whole internet on demand is punishingly expensive. Styskin is blunt about the cost of the index itself: “Don’t ask, it is painfully expensive.” His answer is specialization. “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 said. “That’s why you need to innovate on how you can narrow the search space based on your query very fast.”
The company is also building proprietary retrieval on top of the index. An upcoming product called Web Query Language is meant to help AI systems answer questions by stitching together information from multiple web sources, even when no single page holds the full answer. Keenable recently signed a partnership with voice AI company Gradium to power live information retrieval, though it declined to name its lab customers.
What comes next
Styskin knows the odds. Moving anyone off Google for search is, in his words, “extremely hard.” But he leans on the innovator’s dilemma: Google is potentially “beatable” on agentic queries specifically, where a leaner company can undercut on cost and move faster. The team of 15 engineers across the U.S. and Europe plans to double headcount by year’s end to build out its go-to-market push.
Keenable isn’t alone. Brave and Exa are chasing the same opening, and Google is rebuilding its own search experience for the AI era. That crowd is actually the strongest signal in the story. When the incumbent starts overhauling its core product and multiple funded startups pile into the same lane, it tells you the ground is moving. As TechCrunch AI puts it, whether it’s for humans or for agents, the era of the ten blue links may be closing.
More details are available at the original source.