Databricks on Tuesday announced that it's secured $400 million in new funding, more than doubling the company's valuation to $6.2 billion. Andreessen Horowitz's Late Stage Venture Fund is leading the Series F funding round, which Databricks plans to use to scale its R&D and continue expanding globally.
While the funding round is a significant milestone, it's only the beginning for Databricks, according to CEO and co-founder Ali Ghodsi.
"This is Day Zero for us," Ghodsi said to ZDNet. "Maybe Day One is IPO, but we're not optimizing for that. It's a stepping stone for the vision we have."
Ghodsi envisions his company becoming the platform of choice for companies around the globe that want to add AI to their software.
"We're in a massive massive market," he said. "Think of any software you've used or interacted with -- there are a bunch of things you can make much better by just adding AI... In the next 10 years, every one of those companies behind that software, they're going to add those capabilities or disappear. Every piece of software is going to have much more AI in it."
While Ghodsi said the company will go public at some point, he said there is no particular date set for an IPO. He stressed it's not the company's "end goal."
In addition to Andreessen Horowitz, investors in the Series F funding round included Alkeon Capital Management, funds and accounts managed by BlackRock Inc., Coatue Management, Dragoneer Investment Group, Geodesic, Green Bay Ventures, Microsoft, New Enterprise Associates (NEA), funds and accounts advised by T. Rowe Price Associates, Inc. and Tiger Global Management.
Along with the new funding, Databricks on Tuesday announced that it's hired Dave Conte to serve as Chief Financial Officer, reporting directly to Ghodsi. Conte has more than 30 years of experience in finance and administration within the tech industry. He most recently served as Splunk's CFO, helping to take the company public. Prior to that, he was CFO at Opsware, an IT automation software company that was acquired by HP for $1.65 billion. Conte has also served as a member of the board of directors at Anaplan since 2016.
Databricks was founded in 2013 by the creators of the open source Apache Spark project, the widely-used unified analytics engine for big data processing. The company's Unified Data Analytics Platform (UDAP) is a cloud-based, managed and optimized Spark service available directly from Databricks on Amazon Web Services or as a Microsoft-supported service on Azure. Databricks has thousands of customers around the world, including major enterprises like Comcast, Shell, Expedia and Regeneron Pharmaceuticals.
At a time when many have questioned how and even whether it's possible to build a successful business strategy around open source software, Databricks has thrived doing just that.
Databricks says it is among the fastest-growing enterprise software cloud companies on record. The company's annual recurring revenue has grown more than 2.5X over the past year, it says. And while it brought in almost no revenue nearly four years ago, Databricks' current revenue run rate stands at $200 million.
"We just have crazy demand for our software," Ghodsi said. "We couldn't even forecast this kind of growth a year ago."
That growth is a testament to the company's open source strategy, he said. The cloud-based SAAS business model for open source software is not as well-understood as the alternative strategy, Ghodsi said -- what he characterized as the "Red Hat business model." Offering support services for open source software may be simpler, but customers may quickly lose interest in those services, he explained.
Databricks' strategy, he said, has "much better financial metrics [and is] much more sustainable."
The company's commitment to open source was proven with some recent capabilities added to UDAP: MLflow is an open source framework for managing machine learning experiments and model deployments. Meanwhile, Delta Lake is a storage layer that brings ACID transactions to data lakes. Initially, Delta Lake was a proprietary feature called Databricks Delta. Now, it's an open source project hosted by the Linux Foundation. Ghodsi explained that Databricks only open sources its projects once they're mature enough to gain traction in the market.
With its new funding, Databricks is planning to build dedicated engineering teams to advance Delta Lake and MLflow, as well as Koalas, which allows data scientists to implement the pandas DataFrame API on top of Apache Spark.
The company also plans to invest 100 million Euros over the next three years in its European Development Center in Amsterdam. The European engineering hub has already grown about 3X in two years, Ghodsi said. "It's important for us to have those two" engineering hubs, Ghodsi said, "especially with problems we've had with visas here in the US, it's great to have two locations."
In addition to building out its engineering efforts, Databricks also plans to fund expansion and go-to-market efforts in Europe, the Middle East and Africa, the Asia Pacific region and Latin America.
The biggest challenges for Databricks right now, Ghodsi said, are focusing on execution and maintaining the company's culture and cohesion.
"Growing as fast as we are," he said, it's important "that you keep the culture in tact and keep hiring phenomenal people."
Prior and related coverage: