Building an AI Unicorn: Glean CEO Arvind Jain (Part 2)

Sramana Mitra: This problem has been identified a long time ago. People have tried to build such enterprise knowledge management companies before. I’ve seen many of these attempts. Mostly they’ve failed. While these attempts were made and failing, there was an explosion of digital tools and data. All this was happening in parallel.

In 2019, the enterprise world is more data-rich. There’s more to draw from. AI has become mainstream by this time. What is it in your approach that is different in 2019?

Arvind Jain: You mentioned some of those things like how the world changed. The problems itself became more acute. Now enterprises have so much more knowledge. It’s fragmented. It’s almost impossible to find anything. Whenever you have questions, people will go on Slack to ask where to find the answers. The information was in these hundreds of cloud-based SaaS products. The problem became big.

At the same time, the problem became more practical. One of the biggest issues why no good search products were built in the past was that it was very hard to get hold of enterprise data. These applications have good APIs that allow you to connect and have access to all of that data and to most signals. For example, how are people using content in your company? What is popular? This was not possible in the pre-SaaS world. There were a lot of factors which were just timing-related. We don’t have to worry about how to handle millions of documents.

Then, the core search systems also became available in open domain. We didn’t have to start from scratch. Those are some good enablers. Also, language models are making great progress in 2019. Google had made available a powerful language model to companies like us.

Sramana Mitra: In 2019, you were already building a solution that was language model-based?

Arvind Jain: Right. Language models and semantic search have been part of our product from day one. They’re much better now. Sometimes, there is a big problem that hasn’t been solved. Then a lot of factors come together and that’s what we saw in 2019.

Sramana Mitra: A very important point that I want to underscore is that venture investments in Silicon Valley have been happening at a pretty significant scale since the mid-90s. A lot of ideas were identified as billion-dollar ideas fully-funded as venture-scale thesis. They failed. Given where we are and given the things that have happened, many of those ideas that have failed can now become successful. That’s an interesting way of looking at ideas.

Arvind Jain: As a business opportunity, yes. In the last few years, more capital had become available for startups to build.

Sramana Mitra: Tell me more about that. You have had plenty of startup experience. How did you get your business off the ground? Did you raise venture capital right away?

Arvind Jain: For Glean, we raised venture capital right away.

Sramana Mitra: How much did you raise?

Arvind Jain: We raised a little over $15 million in our first round.

Sramana Mitra: With just a concept?

Arvind Jain: Yes. If you have the experience, you can convince investors that you’re the right person to build it.

Sramana Mitra: That is a crucial point. This opportunity of what you have been able to do of raising $15 million on a concept for a big idea is only available to experienced entrepreneurs with a track record. First-time entrepreneurs don’t have this opportunity.

In the work that we do, there are a lot of first-time entrepreneurs who come to us looking for capital off the bat. That is not available. You have to validate first and then raise capital in more incremental ways. You were able to raise money straight away for, what we call, a fat startup.

This segment is part 2 in the series : Building an AI Unicorn: Glean CEO Arvind Jain
1 2 3 4 5 6 7