- Glean recently raised over $260 million in a Series E funding round at a $4.6 billion valuation.
- Jain previously founded Rubrik, a public cybersecurity company, and was an engineer at Google.
Following OpenAI's latest $6.6 billion fundraise, CEO Sam Altman insisted investors avoid investing in five AI competitors, reported Reuters. Among them are Anthropic, Elon Musk's xAI, OpenAI cofounder Ilya Sutskever's Safe Superintelligence, and AI search startup Perplexity.
The last of the five rivals, and perhaps the least well-known, is Glean, an enterprise search assistant. Founded in 2019 by Rubrik cofounder and ex-Googler Arvind Jain, Glean helps corporate workers find information across their companies' tools and data.
In September, Glean raised over $260 million in a Series E funding round at a valuation of $4.6 billion. Altimeter and DST Global co-led the latest round, which also included Sapphire Ventures, Sequoia, Coatue, and Kleiner Perkins.
Glean helps businesses use AI by tackling a core function for employees—search. "Every company has hundreds, some companies have even more than a thousand different systems or applications," said Jain. "We'll have more and more information over the years, more and more systems." With so much data, it's difficult for employees to find what they're looking for. This can add up to over two hours a day, said Jain.
The company enables AI search by integrating apps like Slack and Dropbox and powering search across their company's universe of data.
Glean hit $50 million ARR over the summer and is projected to end this year with $100 million ARR, according to a source familiar with the company's financials. It is projected to hit $250 million by the end of next year, according to a source familiar. The startup counts Reddit, Pinterest, Sony Electronics, Confluent, Databricks, and Instacart among its customers.
Path to Entrepreneurship
Born and raised in Jaipur, an Indian city known for its historic forts and palaces, Jain was far from the world of technology growing up. "It had very much of a rural feel in the sense that everybody knows everybody in your city, and nobody really moves out."
"I grew up thinking that I would just help my family with whatever they were doing," Jain said.
But in high school, Jain began studying for the entrance exams for the highly competitive Indian Institute of Technology (IIT) schools.
Jain excelled in math and physics, two key subjects for the exams, and as a result, he performed exceptionally well, getting into IIT Delhi. Out of hundreds of thousands of students across India, "his IIT rank was six," said Deedy Das, who led product development at Glean and now works at Menlo Ventures.
At IIT Delhi, Jain studied computer science at the urging of one of the university counselors. "I was interested in pharmaceuticals," he said. But "the counselors were actually quite forceful and told me that I had to study computer science…so it's very accidental." After completing his undergraduate degree, Jain moved to the US to earn a master's in computer science at the University of Washington.
Although engineering was accidental, the desire to become an entrepreneur was deliberate. His father owned "a bunch of small businesses…so he was also an entrepreneur in that sense," said Jain.
In the late 1990s, Jain joined Seattle-based Microsoft and Akamai Technologies, which developed early video products on the Internet. After a stint founding a company, he joined forces with Riverbed Technology, where he served as the founding engineer.
"Nobody was telling me what to do" at a startup, Jain said. "I really got fond of the whole startup experience."
Technical Brilliance
In 2003, Jain joined Google when it was still in the early innings of its growth.
Over the next eleven years, Jain worked on Google Maps, YouTube, and Google Search. Even despite multiple efforts to be recruited to start a company by others, Jain felt inspired during his time at Google—"I felt really happy with the kind of opportunities I got."
Jain was one of Google's first distinguished engineers and reported directly to Google's founding CEO Larry Page, said Das. A distinguished engineer is at level nine out of Google's eleven levels at applies to engineers.
It was only after fellow IIT alum Bipul Sinha came to him with an idea, which ended up becoming Rubrik, that Jain decided to take the leap to start a company. Rubrik, a cybersecurity company focused on data protection and backup, went public in April and is valued at over $7 billion.
"We grew rapidly. We smashed a lot of records in terms of how fast our revenue grew," said Jain. But with that growth came cracks in the team's execution, Jain said. There were challenges to productivity— as a company grows, things inevitably slow down, he said.
Based on an annual survey of Rubrik's employees, one key area for improvement was whether employees had access to the right information to do their jobs. Rubrik had hundreds of cloud-based software-as-a-service apps. "Our company knowledge was sort of fragmented across all these systems," he said.
"Our people are complaining very loudly that I cannot find anything in this company. I don't know where to go and look for things," said Jain.
After realizing there was no product that solved this problem in the market, Jain decided to build it himself. His vision was a search engine that connected a company's data across different apps and made it searchable for employees in a centralized way.
"We had the opportunity to build Google for people in their work lives," and with that, Glean was born.
Glean's Work AI
Jain incubated Glean out of the Kleiner Perkins office in 2019, said Mamoon Hamid, a board member of Glean and a partner at Kleiner. Jain paired up with T.R. Vishwanath, who serves as CTO, Tony Gentilcore, who leads product engineering, and Piyush Prahladka, who led search and is no longer with the company.
"They were in our basement for a good 18 months…I got to see the company really scale from just Arvind," said Hamid.
In the early days, Jain would spend Sundays churning out code, said Das. "He's so happy to get down and dirty…when I read Paul Graham's essay on founder mode, there was nobody else I could think of other than Arvind."
Glean's initial focus was search. "The most basic thing that we'll do with Glean is that you ask the question, it'll surface the right information back to you," said Jain. Enterprise search is a "hard lift and a hard build," said Hamid. "It's seemingly easy, but actually quite hard to build all the different connectors to different SaaS products and then build search itself."
Accomplishing this in a secure and highly-permissioned manner presents a particularly difficult engineering challenge, said Sapphire Ventures' Rajeev Dham. Glean's product, for instance, prevents users from accessing information that they're not meant to see, such as confidential financials or HR reports.
Besides enterprise search, Glean also has an AI assistant that generates answers based on search results, such as summarizing the day's Slack messages or synthesizing multiple documents.
Large language models (LLMs) helped boost Glean's capabilities even further, added Hamid. "We got this massive tailwind of LLMs that we can now infuse into the product and make it truly not just search, but also a work AI assistant." With LLMs, Glean can also generate answers in response to employee queries, including entirely new documents.
Despite Altman urging investors to avoid backing its competitors, Jain remains optimistic. "It's always flattering to see that recognition," said Jain. "The reason why we are likely on that list is because we built something important, something powerful, something that other people aspire to also build."
In the race for enterprise AI, it also helps that Jain is intensely competitive. "Even when you're playing table tennis, ping pong, or tennis, he will fight for that point," said Das. "He doesn't care who you are. You could be a new grad; he doesn't care. He's going to fight for the point."
"I have a very firm belief that companies win or lose because of themselves, not because of competition," said Jain.