Exa raises $250M at $2.2B valuation to build search engine for AI agents, not humans

Investing in Exa | Andreessen Horowitz

For the last two decades, search was built for humans. Google's PageRank turned the web into something navigable: a system of links, rankings, and interfaces, designed to help people find the information they wanted. Entire industries emerged around this model, from SEO to advertising to content optimized for human consumption and focused on maximizing clicks and time spent per page. But we're entering a new era for search, built from the ground up around what AI can do. Increasingly, every meaningful AI workflow, from coding agents to research systems to enterprise copilots, depends on external information as a core input. LLMs are frozen in time; a search engine for AI exists precisely to compensate with fresh, long-tail, real-world context. If the underlying data is stale, incomplete, or incorrect, everything downstream breaks. As one leading AI company put it: "In the limit, if we could search 100% of the time, we probably would. It just comes down to GPUs, latency, and cost." It's easy to underestimate how difficult building search for agents is. Basic search is a commodity; simple keyword searches are cheap and easy. But agents don't ask head queries. Their alpha comes from probing long-tail, constantly shifting information. They write complex queries that can span paragraphs long. In some cases, like with low-latency voice AI, agents need results instantly; but in other cases, as with KYC (Know Your Customer), they need to scan millions of pages to synthesize the most comprehensive answers. There's a Pareto frontier in cost, latency and comprehensiveness for every search query, one that traditional search engines were not designed for. We're also marching into a different scale paradigm: agents will search 1,000s of times more than humans will. Exa's CEO and Co-founder Will Bryk puts it simply: "We're organizing the world's knowledge, but this time for AI." Doing that reliably and economically, at agent-scale, is a genuinely hard problem. It requires building and controlling the entire search stack - something only a few companies have ever done. That's why we're excited to lead Exa's Series C, to back their ambition of perfecting web search, and making it ready for the age of intelligent agents. The signal from the market was consistent. Exa excels with the hardest queries - the long tail of high alpha searches where traditional engines fail. It stands out on low latency where it matters, especially time to first token, which matters acutely in user-facing agent flows. What struck us most was the default behavior: developers and agents are reaching for Exa first. As one customer put it, "this is the default for getting agents to do web search now." No ordinary team would take on such a grand vision: building a search engine from scratch for AI. Will has been obsessed with perfect search for years. He and his roommate, and cofounder, Jeffrey Wang built a search engine in their dorm at Harvard a decade ago. Years before ChatGPT or the AI boom really started, Will and Jeff were inspired by the transformer breakthrough and believed deeply that AI will fundamentally change the way we access information. So they set out on the path to build the search engine for a future where agents become the primary consumers of the web. What stands out is not just their technical depth, but their clarity of focus, developer-first instincts, and hustle. Today, Exa is being used by the leading AI companies, from startups on the frontier like Cursor and Cognition to large enterprises like Hubspot, Monday.com and many Fortune 500s. It's also serving hundreds of thousands of developers who are building agents that are reliant on the accuracy and reliability Exa provides. As AI both consumes and produces the internet, the web will become bigger, faster, and noisier. Search will become the compass for agents to navigate it. The first search wars were won by organizing information for people. The next will be won by organizing information for agents. We're thrilled to partner with Will, Jeff, and the Exa team as they build the perfect search engine for agents and usher in a world of abundant information for all.

Exa Labs raises $250M at $2.2B valuation for its AI search tools - SiliconANGLE

Search startup Exa Labs Inc. today announced that it has raised $250 million in funding to purchase more infrastructure. The round was led by Andersen Horowitz. It comes less than a year after Exa's previous raise, a $85 million Series B investment that included contributions from Nvidia Corp. and Y Combinator. The company is now valued at $2.2 billion. Exa offers a suite of search services that artificial intelligence applications can use to browse the web. The fastest tool in the lineup, Exa Instant, takes under 180 milliseconds to complete queries. The company claims that it's the speediest search service of its kind on the market. Exa stores some of the web data that its services fetch for users in a custom vector database. According to the company, the system can query billions of embeddings in one tenth of a second while using less memory than a high-end personal computer. One of the contributors to the database's speed is that it stores certain important files in the cache of central processing units instead of RAM. Exa ingests public web data using another custom software platform called exa-d. It parallelizes key data processing tasks across multiple graphics cards, which makes it possible to perform them all at once instead of one after another. Additionally, exa-a skips many of the unnecessary file changes that data management systems often make when updating records. That approach further optimizes hardware utilization. Exa's search services turn web data into embeddings, mathematical structures that AI models can understand, before querying them. The company performs the task using custom neural networks called embedding models. It trains the algorithms on an in-house cluster of Nvidia graphics cards. Exa offers its speed-optimized Exa Instant service alongside several other search tools. An offering called Contents enables AI applications to retrieve the full text of webpages. Another service, Exa Agent, facilitates multi-step search workflows. A market research agent, for example, could use it to find an e-commerce store's most popular products and then enrich the list with customer feedback from a reviews site. Exa says that its services have been adopted by more than 400,000 developers. Many of those users work at major tech firms such as HubSpot Inc. and venture-backed startups. The company will use its funding round to expand its AI infrastructure. According to Extra, the hardware upgrades will enable it to train new AI models and process hundreds of thousands of searches per second. The software maker also plans to hire more go-to-market professionals.

Exa Raises $250 Million for AI-Powered Search Infrastructure | PYMNTS.com

The company reached that valuation with a $250 million Series C funding round announced Wednesday (May 20). It is money that Exa says will help it train its next generation of models and scale its systems to support hundreds of thousands of searches per second. "As trillions of agents come online over the coming years, search needs will grow thousands of times beyond the total search volume of Google," Exa Co-Founder and CEO Will Bryk said in the company's announcement. "And as agents make increasingly important business decisions, their requirements for comprehensiveness, freshness, and precision will far exceed what humans require. In short, agents will need perfect search over all the world's information at an unprecedented scale." Exa says that -- unlike traditional search engines designed for human users, Exa provides a web search API specifically optimized for AI products. The company operates its own independent search engine rather than acting as a "wrapper" for existing providers. "As we scale up infra and model training in the coming months, the gap between Exa and wrappers will become clearer," Bryk added. "For example, six months ago we were worse than Google at code search, and now we're used by nearly every coding agent." Since launching its AI-focused API in early 2023, Exa's customer base has grown to more than 5,000 companies, including Cursor, Cognition, HubSpot, OpenRouter, and Monday.com. The funding announcement comes one day after Google unveiled what it called the most significant update to its Search function in 25 years. "Every product search, price check and restaurant booking has started the same way since 1998: Type something short, get a list of links," PYMNTS wrote. "That model is 25 years old," the report added, and Google has announced "it's done with it." In its place is a redesigned interface that accepts text, images, documents, video and open browser tabs and replies with synthesized answers instead of a ranked list of links. In tandem, Google debuted persistent AI agents in Search that monitor topics and push notifications without needing to be prompted. The company said that -- one year after its debut -- AI Mode has topped 1 billion monthly users, queries more than doubling each quarter. Liz Reid, the tech giant's VP of Search, told reporters query volume hit an all-time high last quarter. Google Search and advertising revenue came to $60.4 billion for the first quarter, rising 19% year over year. That went against predictions AI-generated answers would cannibalize Search's ad business.