Exa Raises $22MM to Build the Search Engine for AI

Funding led by Lightspeed Venture Partners will allow Exa to scale their first search product and become the data layer for AI applications.

Exa, an AI research lab redesigning search for the AI age, announced $22M in seed and Series A funding led by Lightspeed Venture Partners, with participation from NVentures, NVIDIA’s venture capital arm, and Y Combinator. This investment will accelerate Exa’s mission to build the search engine for AI.

“Soon, AI will search the web more than humans,” said Exa CEO, Will Bryk. “But search engines like Google were designed for humans, not AI. Whereas Google is optimized for human clicks, AI needs a search engine that’s powerful and precise enough to retrieve thousands of results with the best information. That’s where Exa comes in – we’re the first search engine built for AI.”

Exa trains embedding models, using the same technology behind ChatGPT, to convert web pages into lists of numbers known as embeddings. The result is a technology that packs the power of large language models (LLMs) into the search process itself, making search smarter than keyword approaches like Google. Smarter search grounds AI applications in the most relevant world knowledge.

For example, on Google, a search for “companies in SF building futuristic hardware” returns articles created by search engine optimization (SEO) experts to attract human clicks. On Exa, the same search returns a list of companies that match that description – what was actually asked for.

Also Read: AI and Social Media: What Should Social Media Users Understand About Algorithms?

“Exa represents the intersection of an incredible team, and a big vision for how AI applications will retrieve fresh knowledge. It’s impressive to see what Exa was able to build with such a small team and minimal resources,” said Guru Chahal, Partner at Lightspeed. “The three critical components of AI systems are compute, models, and data. Nvidia supplies the compute substrate. Anthropic, OpenAI and other foundation model companies train the models, and Exa can provide the critical data and knowledge layers. We’re thrilled to support them as they redefine how AI utilizes knowledge and ultimately search as a whole.”

So far, thousands of companies and developers have integrated Exa, from AI writing assistants helping students cite relevant papers, to VC firms sourcing highly specific startups, to AI research teams at companies like Databricks assembling large, high quality training datasets.

“My cofounder Jeff and I actually built a search engine together when we were roommates at Harvard,” said Will. “At the time, we thought crowdsourcing links would enable better search than Google. But now five years later, AI enables something much bigger. AI has the capacity to truly organize the web’s knowledge, and when we do that there will be many magical use cases beyond just a search API.”