Sridhar Ramaswamy says about half of the addressable Global 2000 companies, meaning those outside China, are Snowflake customers.
“half the uh global 2,000 companies that are addressable, uh, that is non-China companies, are uh, our customers”
Person
Sridhar Ramaswamy
4 episodes · 79 quoted claims
- 17 Jun 2026Snowflake CEO: Scaling Data, AI Agents and the New Software Era30m
- 6 Nov 2025Meet Snowflake Intelligence: A Personalized Enterprise Intelligence Agent with Sridhar Ramaswamy42m
- 1 Apr 2025From data breach scandal to AI darling, with Snowflake’s CEO Sridhar Ramaswamy35m
- 10 Oct 2024Snowflake’s CEO Sridhar Ramaswamy on 700+ LLM enterprise use cases55m
Sridhar Ramaswamy says Snowflake operates out of more than twenty-five countries and has customers in many more.
“we operate out of more than twenty-five countries and have customers in way more than those.”
Sridhar Ramaswamy says Snowflake has over 13,000 customers, which helps it predict demand and smooth out spiky usage.
“we are very good at predicting demand because we have over 13,000 customers.”
Sridhar Ramaswamy says customers pay in advance, but Snowflake recognizes revenue only as customers actually use the product.
“You pay first, we recognize revenue when you and your team use Snowflake.”
Sridhar Ramaswamy describes Snowflake Intelligence as an agent-based interface that lets users ask questions of their data in plain language.
“It is roughly an agentic interface to your data, where you can say, how are my biggest investments doing?”
Sridhar Ramaswamy describes MCP as a way for a language model or coding agent to communicate with a data source, acting as an interoperability layer.
“It is just a way for uh, a language model or a coding agent to be able to talk to a data source.”
Sridhar Ramaswamy says Snowflake is developing agent-driven migrations intended to move data from legacy systems in days or weeks rather than quarters or years.
“We are working on things like agent driven migrations, where you can move data from legacy systems onto Snowflake in a matter of days and small number of weeks”
Sridhar Ramaswamy says he succeeded Frank Slootman as Snowflake CEO, and that on his first day the company guided five percentage points below consensus.
“Of course, it does not help that an unknown person that has never been C E O before was taking over from the legendary Frank Slootman.”
Sridhar Ramaswamy says Snowflake launched Cortex Code, a coding agent, as a decision he regards as central to the company's future.
“we should actually launch Cortex Code, a coding agent, because that is going to ensure our future.”
Sridhar Ramaswamy says he started Neva out of a belief that building a better search engine would be a great business.
“Neva was, uh, started out of this sincere belief that creating a better search engine was going to be a great business.”
Sridhar Ramaswamy says technology built at Neva eventually became the foundation of Snowflake's AI work.
“We built amazing technology, and it eventually ended up being the underpinnings of what A I became at Snowflake.”
Sridhar Ramaswamy credits the GDPR with giving consumers the right to demand that a company delete everything it knows about them.
“The fact that you as a consumer can go to a company and say, you need to delete everything that you know about me. That came as a result of G D P R.”
Sridhar Ramaswamy says Snowflake's weekly war rooms gathered everyone responsible for a new product area in one place to tighten the feedback loop between creators and users.
“They brought everybody that was responsible for a new area together into one place.”
Sridhar Ramaswamy says his 24-year-old son, a software engineer at an AI lab, told him that what he learned in school is useless for his current job.
“he told me recently that everything I learned in school and after that is completely useless to succeed in my current job.”
Sridhar Ramaswamy says neither of his parents attended college and both finished only high school, though they strongly valued education.
“Neither of my parents went to college. They only finished, uh, they only finished high school.”
Sridhar Ramaswamy says he attended a college 300 miles from home that his parents were initially hesitant about, but they supported his choice.
“I went to a college that they did not want me to go to, um, because, uh, the idea of sending a young son to a college that was 300 miles away”
Snowflake's original product was conceived more than ten years before this interview by its founder Benoit Dageville.
“Snowflake has always been an amazing product company. The original product that Benoit Thierry conceived of ten plus years ago was many years ahead of its time.”
Ramaswamy says Snowflake's IPO was the biggest software IPO up to that time.
“they had the storied IPO, the biggest software IPO of that at that time.”
Ramaswamy says Frank Slootman voluntarily pushed for a CEO change because he wanted a product-first leader to take charge of Snowflake.
“Frank voluntarily pushed for the change.”
Before Ramaswamy's reorganization, Snowflake had seven to ten layers of teams between an engineer building a feature and the customer using it.
“there were like seven to ten layers of teams that were involved.”
Snowflake created a mixture-of-experts foundation model early in the prior year before abandoning that path.
“Early last year, we, uh, actually went down the path of creating foundation models. We created a credible M o E model.”
Snowflake stopped pursuing foundation models because it lacked the capital to compete with OpenAI and Anthropic.
“We simply did not have the capital to be able to invest meaningfully in things in things like that.”
Ramaswamy estimates that roughly half of qualifying Fortune 2000 companies are Snowflake customers.
“something like half of the qualifying fortune 2,000 companies on the planet are Snowflake customers.”
Snowflake built an internal sales data assistant called Raven and then worked with early customers including Cisco, Fanatics and the USA bobsled team on Snowflake Intelligence.
“But it's basically the sales data assistant. Then we started working with early customers, whether it is a Cisco or a fanatics, or the USA bobsled team”
Snowflake Intelligence is priced on consumption, so customers pay for what they use.
“Snowflake intelligence is very much a consumption product. People pay for what they consume”
Snowflake rolled out coding agents to its solution engineers, which sharply reduced the time needed to create customer demos.
“We've rolled out coding agents to all of, uh, our solution engineers, and they are excited because that just dramatically lowered the amount of time it takes to create a demo.”
Snowflake founder Benoit Dageville helped drive adoption of coding agents among Snowflake engineers.
“Benoit, our wonderful founder, who fell in love with, uh, coding agents, did more to drive coding agent adoption with engineers.”
Ramaswamy says his first Google launch, which was his own work over three months, was covered by the New York Times.
“my first, um, launch at Google, which was entirely my work for three months. One person was covered by the New York Times.”
Ramaswamy says the three hyperscale cloud providers would like to own the data space, but Snowflake and Databricks exist as independent successes.
“The three hyperscalers would love to just own the data space like they own any other space, but yet there's Snowflake, there's Databricks”
Ramaswamy says Google largely stopped at information and did not succeed in expanding into physical-world areas like shopping, airline purchases or hotels.
“Google, for example, stopped at information. God knows I spent enough time trying to get into physical things like shopping or airline purchases or hotels.”
Ramaswamy says Snowflake's relationship with Microsoft was initially mediocre because Microsoft had a first-party relationship with Databricks and was conflicted about its Fabric product.
“Among the earliest places where, uh, this went to work was in our relationship with Microsoft, which was okay, but not that great.”
Snowflake is deepening its partnership with SAP, including bidirectional data sharing and joint work on analytics, AI and agents.
“So with folks like SAP, we are actually thinking much harder about what is that one plus one equals three with SAP.”
Ramaswamy says he left Google about seven years before this interview and is now more proud of the work his search ads team did.
“I'm actually more proud of the work that we did in the search ads team now than I was when I left Google.”
Ramaswamy's startup Neva launched citations in its search product in early 2023.
“how proud we were then that we launched it in, uh, in early 2023.”
Ramaswamy says PageRank was the insight that powered Google, using the whole internet to determine what was popular.
“The insight that powered Google was PageRank. It was a way of harnessing the power of the entire internet to figure out what was popular and what was not.”
Ramaswamy says PageRank's usefulness faded by about 2004 or 2005, after which click behavior on search results became more important to Google.
“But PageRank ran out of juice, like in six years, like 2004, five.”
Sridhar Ramaswamy says about 700 Global 2000 companies use Snowflake.
“We have something like 700 odd global 2,000 companies that are on Snowflake.”
Sridhar Ramaswamy says large public companies close their monthly books on Snowflake.
“Large public companies close their books every month on top of Snowflake.”
Sridhar Ramaswamy says Snowflake is central to how data moves within the US financial system.
“Snowflake is the beating heart of the, at least the US. financial system in terms of how data moves from place to place.”
Sridhar Ramaswamy says Snowflake hosted the DeepSeek model very quickly after its release.
“Things like hosting DeepSeek quickly, you know, that's just a little bit of making sure that you can still run the hundred meter sprint in ten seconds.”