Today we are talking about one of the most exciting AI companies in the enterprise tech space. Arvind Jain started Glean in 2019 before ChatGPT and before the AI boom. Why? Because he was frustrated that he couldn't search for data at his own earlier company Rubrik, which he had co-founded. Fast forward to today. Glean crossed $100 million in ARR early this year and they are doubling it. They are building what Arvind calls a better version of ChatGPT for enterprises. An AI assistant that actually knows your company's data, not just the internet's. With over 1,100 employees, Glean is seeing global tech giants copy what Arvind and his team did. In the latest episode of ET in the Valley, he tells me what it's really like to create an entirely new category, why being conservative was his biggest mistake, and why he has never been more optimistic. It's Wednesday, the 31st of December. I'm Swati Morti. This is the Morning Brief.
Hi Arvind, great to have you with us. You started Glean in 2019.
And you had crossed 100 million ARR in early 2025. As you head towards the 200 million mark, what surprised you and worried you the most in this journey?
Well, today, the business is growing quite rapidly. We have more than doubled from where we ended our last year and I think we see a big growth ahead of us as well. In terms of challenges, we've gone through the usual challenge that any other startup goes through. When we started, we built a product but nobody knew about us and nobody in fact was used to buying a product that we were selling. We were building something that they never buy. They don't buy a thing like that. And so then you have to actually create extra dollars, extra budgets to make room for a new technology purchase like ours. In a market like that, when you're defining a new category, it takes time to actually get those initial customers. But once you achieve some success, you're also known as a pioneer and a category creator, which is what today we're known as. You will see oftentimes in media people talking about how so many other large companies are coming and trying to build a product like Glean because we got that recognition as being the category creator, as being the first to bring transformers and generative AI into the enterprises. So I think now the path is different for us. Now we don't have to convince our customers to actually buy a product like ours, in the sense that they already know that they need something like Glean, and that allows us to actually go and scale faster.
You spent a decade at Google and then you co-founded Rubrik. What made you start again and why this problem in specific?
Yeah, Glean was very easy as a product that we wanted to build because it was for us. I personally, in every one of my jobs before, really struggled to have answers to questions that I have, and whenever I needed to find something inside the company, it would be a nightmare. You could never find anything internally. Sometimes you'd find documents but they look out of date and stale, and most of the times you don't find anything. In Rubrik, we grew very fast. With that fast growth, actually ultimately we came to a place where our productivity came to a screeching halt. We were not able to release our software. So we would ask, what is happening in the company? What are the problems? And one of the biggest things that people complained about was not being able to find things inside the company or not being able to get help from people with their tasks. As an employee, for me to get my job done, I need to have quick access to information. I need to have access to people who can help me do my job. This was a big problem for us. We were scoring very low in our pulse survey on that question. That problem was very relatable to me, and so that's sort of what led to the creation of Glean. It is also coming back to my roots because I worked at Google for a long time and I worked on Google Search. I had many of those teams, and search is always an exciting problem for me. And this time, we had to solve the same problem but in a different context, which is inside a business.
Would you say your life has come a full circle?
Certainly. And I think search is most exciting to me because it's like magic. For a long time, Google was like magic. You could actually type in one or two words and you can misspell those words, but somehow Google would read your mind and bring that information back to you. And there are a lot of amazing things that we did at Google which were magical. And the thing as a developer of that technology, the most fun thing was that you are a user of that product yourself. Every day internally when I was working at Google, we would ask questions in Google all the time, and sometimes we'd get the answers back, sometimes you won't. But when we didn't, my curiosity would go up and I would try to go back and see why that is. Why is Google not surfacing the information I want at the top? And so it's sort of like when you use the products and they don't work well, then you get that drive to actually go and make them better. If you feel like you can do it, right? So that's the thing that I really like about search, is that I'm building a product for ourselves.
Have you actually solved the problem you set out to solve when you started Glean?
I mean, I think we have solved a good part of it. We built a really good product. We have actually solved even more than what we thought. Glean is the first enterprise generative AI company in the world.
To the best of our knowledge, we started in early 2019. At that time, language models were not really the talk of the town. They were not that popular. In fact, initially this technology was built at Google to make Google Search better. But at the time, none of us actually knew how good these models were going to become in the next few years. Our sort of vision only went as far as that it'll allow us to build a great search product. We at that time did not think about that machines will be able to write like humans. We actually at that point didn't feel like machines could actually also reason at a level of a human, and now they've gotten close to that, sometimes even better. You're not just a Google-like search experience inside your company. Today, think of it as a better version of ChatGPT. It's an AI assistant. You go to Glean, you ask it any questions or you give it some tasks, complex tasks to do, and it's going to answer those questions or complete those tasks using all of the world's knowledge and wisdom, but also using all of your internal company's data, knowledge, information, and context. In that sense, Glean becomes that one-stop shop, one place where you can go and you can get a lot of work done. At Glean, it is a place where AI technology comes together with your company's context to help you with your day-to-day tasks.
Arvind, you were an early mover in the enterprise AI space, but now it's crowded with companies like Elastic and other enterprises building in-house tools. Do you worry about losing market share? And more importantly, is being first still an advantage in the AI landscape currently?
We're too early to lose market share. We are building our business, and so we are more than doubling or tripling our business every year. So we are in the buildout phase. There is a lot of competition. There are a lot of people who have come and solved the problem the way we do. But as you said, we are the pioneers and we have that recognition for being the best product in the market by far. For us to succeed in a business, we always have to make sure that we build great technology, but then we also make it very accessible. Especially, for example, in the India market, we want to make sure we're not actually very deeply penetrated in the India market yet. That's a big goal for us now. We have a big presence in India from the beginning, but now we've actually also aggressively building our sales team, and we will make sure that it's actually priced and it's affordable. Because I think our goal is to first add value. That's sort of the way we think about our business.
You mentioned India earlier. How big is your team in the country right now?
We probably have more people. We do a lot of our core technology development, R&D, and we have a large office in Bangalore. We must have like at least 300 people there.
Oh, does it mean at least a quarter of your workforce is in India?
Yeah, maybe. We've been growing fast. So in fact, I don't have the latest numbers, but the last I checked, it was I think about 1,100, it was north of 1,100. And I think we do have about 300 people in Bangalore. We built Glean with that in many ways. For example, for our core technology development, we wanted this model of equal contribution from both of these locations, from Silicon Valley and Bangalore. We did the same at Rubrik, and I did the same at Google. I ran Google India and I established our R&D presence there early on.
Where does India stand from a revenue and growth perspective?
See, India is one of the largest economies in the world, and there are a lot of amazing large enterprises, which is what our focus is. We sell to large businesses generally. We also sell to midsize, and we hope to grow our business probably like 10 times next year and then the same the year after, because starting with such a small base, huge growth is what we're going to be looking for next year in India.
How are the buying dynamics and demand patterns changing? Because if you look at enterprises, their willingness to pay has always been low, and the current explosion of AI products and subscriptions aren't really helping. Are you seeing genuine willingness to pay?
There is a lot of interest for AI technologies in India. The Indian companies are smart. They know that they have to invest in AI and that it could actually bring a lot of higher profits and higher capacity to do more, increase your top line. So people understand it. Therefore, no, I don't actually see it being any behind than any other market in terms of the cost and all that. Companies are very eager to capture market share. The pricing is in some ways proportional to value, but it also sometimes is even less than that because it also has to be somewhat related to what is your cost of building it. I like to think of technology like that. I myself like to buy technology which is fairly priced, and so we won't actually go and not do the same ourselves.
Arvind, you've been doubling and sometimes even tripling your revenue year after year. At the same time, there is a growing talk in Silicon Valley about an AI bubble. Do you expect this momentum to sustain, and how are you preparing for that risk?
