Synopsis
Snowflake is prioritizing tightly integrated global engineering teams for core product development, impacting its R&D expansion plans in India due to timing and coordination challenges. While India is a growing market and ecosystem hub, the company emphasizes making AI usable across entire organizations, not just isolated pilots.Listen to this article in summarized format
“There is no firm plan for a full R&D setup in India. The issue is not talent. The issue is timing,” Benoît Dageville, cofounder of AI data cloud company Snowflake, told ET in an interview at the company’s annual summit in San Francisco. He added that core product development requires tightly integrated teams that work in constant coordination, making distributed ownership more complex.
NYSE-listed Snowflake, founded in 2012 by Dageville, Thierry Cruanes, and Marcin Żukowski, holds a market capitalisation of $82.58 billion.
He acknowledged that India is becoming an important part of Snowflake’s growth strategy, both as a market and as an ecosystem hub.
The company has expanded its commercial presence in the country and is also using India as a growing centre for partner development in the Asia-Pacific region.
On the next phase of enterprise AI, he said, it is not about isolated pilots but about making AI usable across entire organisations. “AI is not only about use cases like improving customer support or internal processes. It is about putting AI in the hands of everyone in the organisation,”.
Dageville noted this shift is forcing companies to rethink how data and compute systems are designed. “You have all data in one place, you have all the compute, and the nature of compute needs to be very broad so you can support all workloads,” he said, adding that this includes analytics, AI workloads, application deployment and transactional systems.
As adoption expands, governance is becoming a key requirement. “You don’t want every user to see all the data. There is confidential data. You need governance,” he said.
The debate in enterprise AI is also a lot about where value sits across the stack, and how it is priced in the market. Snowflake has seen strong growth and improved guidance, but valuation concerns remain. Even companies with strong growth, such as Palantir Technologies, have faced pressure as high valuations weigh on stock performance.
Dageville said Snowflake’s focus remains on building the underlying platform. “We are building the stack,” he said.
On the broader AI cycle, Dageville argued the industry is still moving from early experimentation to real production systems. “We are still moving from early experimentation to real production systems, where execution matters more than hype.”
Speaking candidly about the company’s origin, Dageville said the name Snowflake came from both personal interest and a conceptual idea. It reflected his interest in skiing, but also the idea that individual pieces of data, like snowflakes, only become powerful when combined at scale in the cloud.
(The reporter was in San Francisco at the invitation of Snowflake)
