Databricks Data + AI Summit 2024 | Anjan Kundavaram | Fivetran

theCUBE.net

Clips
News
More from Databricks Data + AI Summit 2024

Anjan Kundavaram

Chief Product Officer

Fivetran

Seamless data movement: Fivetran aims to unlock AI’s full potential with efficient data integration

In today’s fast-paced world of data and artificial intelligence, integrating strong data management systems with advanced AI technologies is essential for businesses to remain competitive. As AI continues to gain prominence, the focus on efficient data integration and management has become critical to fully harness AI’s potential.Innovations in data movement and storage are paving the way for more seamless and effective data utilization, thereby driving business agility and insight generation

play_circle_outline Databricks Data and AI Summit in San Francisco

play_circle_outline Unlocking the Power of AI and Analytics: Exploring Fivetran's Seamless Data Movement Solutions

play_circle_outline Growth of data management and AI readiness in the industry

Info

Anjan Kundavaram | Fivetran

Anjan Kundavaram

Chief Product Officer Fivetran

search

>> Good morning data fans and welcome back to stunning San Francisco. It happens to be a fabulous summer day here, which we don't always get to say. We are at Databricks Data and AI Summit. My name's Savannah Peterson, joined by my fabulous and co-analyst today, John Furrier. John, good morning.

>> I mean, it's a great show. Databricks always brings the tech angles that are more open source oriented, and it's a market that's growing and all the picks and shovels are coming out and the tooling's growing. So it's a growth market, but there's a lot to fix, and so products that are winning are the ones that are moving the needle. So this next segment, we're going to get into it.

>> Speaking of products that are winning, really excited to welcome onto the stage, Anjan from Fivetran. Thank you so much for being here.

>> Thank you for having me. Excited to be here.

>> What makes you excited to be here this morning?

>> Look, it's a great time to be in data. AI is like... Everyone's talking about AI, but for you to kind of get to AI, you got to be at data and Databricks is driving a lot of innovation. Some of the key things that you saw this morning are exciting. I mean, they had me at AI and query optimization. AI is making things faster for us, so it's a great keynote I thought.

>> It absolutely is. Fivetran has a new role for you three weeks ago. Congratulations.

>> Thank you.

>> You mentioned how easy Fivetran is to use and that was a big part of the appeal for you joining. Why is it so easy to use?

>> Look, part of the reason when I joined Fivetran is the product is so simple to use. If you want to move data from say Snowflake or Postgres to Databricks, you could do four clicks and you move your data, and that's not easy to do. Underneath, there's a lot of complexity. It's one thing to move data, it's another thing to move data correctly. So all the hard parts, the team does an amazing job fixing it, so it looks super easy on the top, but underneath, okay, there's a ton of stuff the team has to do. Schema, evolution, data, getting there correctly, and this is sort of where DIY, so customers don't want to do it on their own. They kind of get it right, ones. Then data, once the schema changes, the data doesn't do it right? So I'm super excited to be here at Fivetran. Amazing vision by George and Taylor. I love where George has taking the company. So super excited to be at Fivetran.

>> Congratulations. You guys got great rave reviews on theCUBE last week. I want to ask you the question around the data movement. We heard Olly Godsey say on stage, data fragmentation, the data and states are fragmented. Huge problem. Lakes are supposed to solve that, but they still got to work on that. What's the driver of the data movement? Is it the GM customer was saying they had a hunt for all the data, took up all their time. Is it time? Is it they're resetting up the data for AI? Is it because that's the natural fluid nature of it for harmonization?

>> Yeah, so if you're a company, you've got tens of sources, right? You've got lots of databases, lots of applications, and then for analytics you're like, okay, I'll just get the operational data out of the databases. I'll get some of my applications. But now with AI, oh, some of your context is in these other applications that are super important. So you really need to get the data to a single destination or a storage layer. If you don't do that, you're missing out. So I think that's the first motivation is if you want agility, if you want to go really run analytics and then LLMs are able to do things analytics coding. They're able to parse their unstructured data. They're able to parse through sources and give you new insight. So I think it's paramount companies first invest in moving data and moving data quickly, and if you're a company that is looking at stale data from one of these sources, that could be the missing thing for driving a new business model, driving innovation. So I think part of this is like it's not being a focus for companies and analytics should always be important to put the data out there. I think AI makes it imperative to do it. And then a tool like Fivetran, we make it so easy for you. I got a quote from a customer the other day. It would take them six weeks to do one source to a destination, and he told him, he got on a meeting-

>> Six weeks?

>> He got on a meeting with his team and did it live on a call, live. It just moved the data.

>> That's how it should be.

>> And that's our job. We are the plumbing company. We're going to make it super easy to move the data and we do it across 500 data sources and we've invested a ton there. I can shout a little bit of how we're doing that.

>> Real quick, just so people know how they can consume it, on-prem, in the cloud, hybrid. What's the consumption?

>> We've got a few different models. You can certainly just literally go sign up five minutes, move data. We're a SaaS company. We've recently launched a hybrid deployment. Customers can go add these things in our SaaS control plane, but the data plane sits on the customer site, so it could be on-prem, could be VPC. So for most companies that are security conscious, that model will do.

>> You're a plumber, data plumber, I like that term. Who's your target customer and does is change with GenAI? Our keynote review this morning, we talked about how Databricks are looking at their personas. It's shifting significantly. Data engineering is still hot category. Who do you guys talk to? Are you more data infrastructure, data engineering?

>> I think it is a spectrum. Certainly data engineers, data teams, they'll be like, "Hey, I don't want to do this anymore. I'm going to do something more valuable." If you're a team that's kind of more on the analyst side, like, oh, we've got a great product for you. We have LOB triers out. So I think it's a spectrum. If you're an analyst, you can use our product. If you're an engineer, you can use our product. It's kind of built for the scale and the throughput for an engineer needs, and it's built for the simplicity for an analyst.

>> So it's going back to that easy to use piece that you were talking about across the organization. I love that. So talking about your customer types, we've heard we talked about Lira Health on Silicon Angle last week. I know you've got a couple examples for us today. Maybe give us a little Conway, maybe a little Goldman Sachs,

>> Yeah. Yeah. So Conway and I asked there, it's super interesting. I was reading the notes on this one. They're paying a company of GQ and Vogue and so on. They have list of 500 million active users. So online they've got close to a hundred million print users.

>> I mean, every single one of us probably reads something that Conway publishes every day, whether we realize it or not.

>> But think about it, they generate trillions of data points.

>> Absolutely.

>> Like you and me reading it, how am I using it? How many magazines are we using it? So they weren't able to collate all of that in a single place. So we helped them put all of the data in Delta in Databricks, and now you've got all the data and you can kind of do, oh, I can give you personalized content. I can tell you what to do. That likely drove new business models and new monetization opportunities. So it started with let's collect all the data in time that actually in a way that makes a difference for them. So it's been a huge success story, being a great partner driving that story.

>> I think that's awesome. I'm personally a Conway consumer, so I think about how-

>> So we're helping you get better content.

>> I'm thinking about how many different data points that I pull from them if I browse a GQ article versus read all of the travel stuff I'm looking at for this summer. It's funny, you mentioned you also had a Goldman Sachs AI example.

>> Yeah, let me shout about Saks actually. Just like the retailer Saks.

>> Oh, Saks. Sorry.

>> Yeah, so this is an interesting story as well. So there's a little bit of an AI angle there. Again, same thing, like struggling. They have lots of data sources, and the first problem is like, "Hey, why don't we help you put the data in a data lake and start thinking about it?" But what they're doing now is okay, now they have all that context. So if they get a phone call and customer support sort of picks up. Now they've made the investment to put all the data in a single location, they can point to an LLM and the LLM's kind of giving them an assist. Every time you call them and ask them a question, they're getting a kind of a packaged answer, and it's up to the customer service agent whether they want to use that or not. So this is where all that data investment, now they're able to turbocharge with AI, which I think is fascinating.

>> And enhance that customer experience.

>> Exactly.

>> I love it. I want to ask you about the relationship you guys have with Databricks. Obviously it's their show. You guys are in the ecosystem, Ali's on stage saying ecosystem's critical to their success. Obviously it's a platform, they need it. Although Jensen Wong called the DIP, he is almost as good as NIMS, which is his acronym, but they've got the platform. You guys are a partner. How about the relationship, how you guys work with Databricks? How's that going?

>> We are a very close collaborator with Databricks. We have been for a number of years. I think the best way to kind of count that is the number of joint customers we have. We have hundreds of joint customers with Databricks. We work very closely, kind of excited about some of their announcements. There's something that'll come up tomorrow that we're working with them on. So it's been a very deep relationship with Databricks. George and Ali talk all the time, and I think it's really the foundation of the partnership has joint customers.

>> Keep those lakes clean, right?

>> Keep those lakes clean, and I kind of want to commend what they've done. They're committed to being open and also kind of Snowflake, I'd say super eager to see how that kind of pans out.

>> I mean, open-sourcing Unity was a pretty big deal and it was kind of already open, but that gives the governance-

>> Kudos to Polaris and Snowflake for open-sourcing, kudos to Ali for open-sourcing Unity. It is a great thing for an industry. Let's figure out what customers want and then what's in there. There's no more black boxes.

>> From a product perspective, share your thoughts on open-source relative to developers standardizing around or formatting with data, because as data becomes generative, it's going to be in line in the coding cycle. So data is going to be part of coding where you're starting to see that now, not just in database. That part decoupling compute from data is going to enable some pretty cool things. What's the open-source angle? What's in it for the open-source participants? What's the ideal product scenario to evolve, if you had to guess that?

>> No, look, I think first you're going to get a ton of adoption when things are open. We've seen so many examples in our industry, Spark or Kafka and so on, and there are competing projects that didn't take off. You get overnight a ton of developers that are going to go commit and so on. You get a ton of free developers building, and then it still leaves vendors and opportunity to go and build the enterprise capabilities, kind of go innovate on top of that. But you are partnering with the open-source ecosystem. Where it goes wrong, is you try to do something that's against the customers, let the customers drive you, and as a product guy, you really want to go where the customers want. And the best way is go talk to a bunch of customers using the open-source and be like, what are the problems you're looking at and help solve that. So it becomes a lot easier as a product guy.

>> And what's your view on Iceberg and Delta Lake?

>> So I think they are two great table formats. I think we support both of them. Exciting to see how maybe both of them come together. I think the thing that I love is we are finally talking about a couple of formats, maybe one or two catalog formats as well. And we are going to spend... Actually, we've been spending a ton of time optimizing and moving data to Iceberg and Delta. So it's finally good to see open table formats, open metadata catalogs, and then we're going to be right there with our customers.

>> I'm curious, because I know the role of a chief product officer, there's a little bit of tension with marketing and some of the other teams. As you prioritize what's in your product roadmap, how do you approach that? It's a big challenge when things are moving as fast as they are right now.

>> No, I think it's a really good question. And often I think about quantitative and qualitative data, and if you go interview your best customers and maybe sort of thought leaders and you get a ton of data on how are customers using it, maybe some open source customers like we mentioned, and then you put that in front of... You sort of massage the data and then kind of how you prioritize becomes a boring exercise. But if you don't have the data, then it kind gets into a conflict. So Rachel and I, yeah, because in the data she's like, yeah, I think we should do that. Right? And then obviously there are strategic bets that you got to make and then do that with conviction.

>> You know your data is good if that conversation is boring. I really like that actually as a benchmark, I don't think I've ever had one of those conversations be boring. So I think that's a really. Well, obviously, your process is working, if you're figuring that. Do you feel like... I mean, you've been in data and analytics for a long time. It feels like the light is shining on your space a little bit more right now than normal. Would you agree? Is this a moment for you?

>> No, it's a great time for us. So if you think about, we've got many sort of tailwinds, AI. For you to build great models and great insight, the first thing you have to do is move data. We've invested in Iceberg and Delta, and this is important. We have 500 data sources that you can click a button and you get an Iceberg table or you get a Delta table. And that for a customer who's now thinking about that is less headache. It's hard to move that data when the schema changes, how do you think about updates? How do you think about governance? So we're just going to take all of those problems away for customers. So we feel like we've got two great trends, we've got great customers, and we've got over 200 customers using our Managed Data Lakes product. That's like, where are the pucks going?

>> What's the big drivers on the product-led growth, you guys are seeing, obviously you got a great product. Congratulations. I mean, having a simple product that's easy to use, that does a lot of high-performance capabilities is a winner. We've seen that. Is it the AI readiness or is it just data management in general's still growing? More data's coming in?

>> Yeah.

>> Or both. How would you rank the-

>> Actually, it's funny. We'd be surprised to find data's even growing on-prem. Data's going in the cloud. Data's going everywhere. We're just collecting more data, which is good for us, right? I think on the product-led growth, if a customer likes it and they're going to tell their LOB, they're going to tell their colleagues, it's really that. If the experience is gray, then the PLG sort of life cycle just go through. It's like a flywheel, we think. So really obsess about moving data, moving data correctly, moving to the right formats, and sort of tweaking all the knobs and you're like, "Hey, I just need to turn the button on. It works."

>> Everyone wants that easy start button.

>> Yeah.

>> All right. It's a very exciting time. Obviously you probably have a little... You mentioned there's an announcement coming tomorrow. I won't let you spoil it for us, but more broadly speaking, the next time we have your lovely self on our show, let's say at the summit next year, what do you hope to be able to say then that you can't say today?

>> I think I'd hope our customers would say if someone's moving data to Iceberg or Delta, they're like, "I'm just using ."

>> Perfect. Love the sound bite. We will be playing that back on our highlight reel when we have you back on the show. Thank you so much for taking the time to be here with us.

>> Absolutely.

>> Such an exciting.

>> Thanks for having me.

>> Hey, our pleasure, John. Great questions as always, and thank all of you for tuning into our two days of fantastic coverage here in our hometown of San Francisco, California. My name's Savannah Peterson here at Databricks Data and AI Summit. You're watching theCUBE, the leading source for enterprise tech news.

Anjan Kundavaram | Fivetran

search

>> Good morning data fans and welcome back to stunning San Francisco. It happens to be a fabulous summer day here, which we don't always get to say. We are at Databricks Data and AI Summit. My name's Savannah Peterson, joined by my fabulous and co-analyst today, John Furrier. John, good morning.

>> I mean, it's a great show. Databricks always brings the tech angles that are more open source oriented, and it's a market that's growing and all the picks and shovels are coming out and the tooling's growing. So it's a growth market, but there's a lot to fix, and so products that are winning are the ones that are moving the needle. So this next segment, we're going to get into it.

>> Speaking of products that are winning, really excited to welcome onto the stage, Anjan from Fivetran. Thank you so much for being here.

>> Thank you for having me. Excited to be here.

>> What makes you excited to be here this morning?

>> Look, it's a great time to be in data. AI is like... Everyone's talking about AI, but for you to kind of get to AI, you got to be at data and Databricks is driving a lot of innovation. Some of the key things that you saw this morning are exciting. I mean, they had me at AI and query optimization. AI is making things faster for us, so it's a great keynote I thought.

>> It absolutely is. Fivetran has a new role for you three weeks ago. Congratulations.

>> Thank you.

>> You mentioned how easy Fivetran is to use and that was a big part of the appeal for you joining. Why is it so easy to use?

>> Look, part of the reason when I joined Fivetran is the product is so simple to use. If you want to move data from say Snowflake or Postgres to Databricks, you could do four clicks and you move your data, and that's not easy to do. Underneath, there's a lot of complexity. It's one thing to move data, it's another thing to move data correctly. So all the hard parts, the team does an amazing job fixing it, so it looks super easy on the top, but underneath, okay, there's a ton of stuff the team has to do. Schema, evolution, data, getting there correctly, and this is sort of where DIY, so customers don't want to do it on their own. They kind of get it right, ones. Then data, once the schema changes, the data doesn't do it right? So I'm super excited to be here at Fivetran. Amazing vision by George and Taylor. I love where George has taking the company. So super excited to be at Fivetran.

>> Congratulations. You guys got great rave reviews on theCUBE last week. I want to ask you the question around the data movement. We heard Olly Godsey say on stage, data fragmentation, the data and states are fragmented. Huge problem. Lakes are supposed to solve that, but they still got to work on that. What's the driver of the data movement? Is it the GM customer was saying they had a hunt for all the data, took up all their time. Is it time? Is it they're resetting up the data for AI? Is it because that's the natural fluid nature of it for harmonization?

>> Yeah, so if you're a company, you've got tens of sources, right? You've got lots of databases, lots of applications, and then for analytics you're like, okay, I'll just get the operational data out of the databases. I'll get some of my applications. But now with AI, oh, some of your context is in these other applications that are super important. So you really need to get the data to a single destination or a storage layer. If you don't do that, you're missing out. So I think that's the first motivation is if you want agility, if you want to go really run analytics and then LLMs are able to do things analytics coding. They're able to parse their unstructured data. They're able to parse through sources and give you new insight. So I think it's paramount companies first invest in moving data and moving data quickly, and if you're a company that is looking at stale data from one of these sources, that could be the missing thing for driving a new business model, driving innovation. So I think part of this is like it's not being a focus for companies and analytics should always be important to put the data out there. I think AI makes it imperative to do it. And then a tool like Fivetran, we make it so easy for you. I got a quote from a customer the other day. It would take them six weeks to do one source to a destination, and he told him, he got on a meeting-

>> Six weeks?

>> He got on a meeting with his team and did it live on a call, live. It just moved the data.

>> That's how it should be.

>> And that's our job. We are the plumbing company. We're going to make it super easy to move the data and we do it across 500 data sources and we've invested a ton there. I can shout a little bit of how we're doing that.

>> Real quick, just so people know how they can consume it, on-prem, in the cloud, hybrid. What's the consumption?

>> We've got a few different models. You can certainly just literally go sign up five minutes, move data. We're a SaaS company. We've recently launched a hybrid deployment. Customers can go add these things in our SaaS control plane, but the data plane sits on the customer site, so it could be on-prem, could be VPC. So for most companies that are security conscious, that model will do.

>> You're a plumber, data plumber, I like that term. Who's your target customer and does is change with GenAI? Our keynote review this morning, we talked about how Databricks are looking at their personas. It's shifting significantly. Data engineering is still hot category. Who do you guys talk to? Are you more data infrastructure, data engineering?

>> I think it is a spectrum. Certainly data engineers, data teams, they'll be like, "Hey, I don't want to do this anymore. I'm going to do something more valuable." If you're a team that's kind of more on the analyst side, like, oh, we've got a great product for you. We have LOB triers out. So I think it's a spectrum. If you're an analyst, you can use our product. If you're an engineer, you can use our product. It's kind of built for the scale and the throughput for an engineer needs, and it's built for the simplicity for an analyst.

>> So it's going back to that easy to use piece that you were talking about across the organization. I love that. So talking about your customer types, we've heard we talked about Lira Health on Silicon Angle last week. I know you've got a couple examples for us today. Maybe give us a little Conway, maybe a little Goldman Sachs,

>> Yeah. Yeah. So Conway and I asked there, it's super interesting. I was reading the notes on this one. They're paying a company of GQ and Vogue and so on. They have list of 500 million active users. So online they've got close to a hundred million print users.

>> I mean, every single one of us probably reads something that Conway publishes every day, whether we realize it or not.

>> But think about it, they generate trillions of data points.

>> Absolutely.

>> Like you and me reading it, how am I using it? How many magazines are we using it? So they weren't able to collate all of that in a single place. So we helped them put all of the data in Delta in Databricks, and now you've got all the data and you can kind of do, oh, I can give you personalized content. I can tell you what to do. That likely drove new business models and new monetization opportunities. So it started with let's collect all the data in time that actually in a way that makes a difference for them. So it's been a huge success story, being a great partner driving that story.

>> I think that's awesome. I'm personally a Conway consumer, so I think about how-

>> So we're helping you get better content.

>> I'm thinking about how many different data points that I pull from them if I browse a GQ article versus read all of the travel stuff I'm looking at for this summer. It's funny, you mentioned you also had a Goldman Sachs AI example.

>> Yeah, let me shout about Saks actually. Just like the retailer Saks.

>> Oh, Saks. Sorry.

>> Yeah, so this is an interesting story as well. So there's a little bit of an AI angle there. Again, same thing, like struggling. They have lots of data sources, and the first problem is like, "Hey, why don't we help you put the data in a data lake and start thinking about it?" But what they're doing now is okay, now they have all that context. So if they get a phone call and customer support sort of picks up. Now they've made the investment to put all the data in a single location, they can point to an LLM and the LLM's kind of giving them an assist. Every time you call them and ask them a question, they're getting a kind of a packaged answer, and it's up to the customer service agent whether they want to use that or not. So this is where all that data investment, now they're able to turbocharge with AI, which I think is fascinating.

>> And enhance that customer experience.

>> Exactly.

>> I love it. I want to ask you about the relationship you guys have with Databricks. Obviously it's their show. You guys are in the ecosystem, Ali's on stage saying ecosystem's critical to their success. Obviously it's a platform, they need it. Although Jensen Wong called the DIP, he is almost as good as NIMS, which is his acronym, but they've got the platform. You guys are a partner. How about the relationship, how you guys work with Databricks? How's that going?

>> We are a very close collaborator with Databricks. We have been for a number of years. I think the best way to kind of count that is the number of joint customers we have. We have hundreds of joint customers with Databricks. We work very closely, kind of excited about some of their announcements. There's something that'll come up tomorrow that we're working with them on. So it's been a very deep relationship with Databricks. George and Ali talk all the time, and I think it's really the foundation of the partnership has joint customers.

>> Keep those lakes clean, right?

>> Keep those lakes clean, and I kind of want to commend what they've done. They're committed to being open and also kind of Snowflake, I'd say super eager to see how that kind of pans out.

>> I mean, open-sourcing Unity was a pretty big deal and it was kind of already open, but that gives the governance-

>> Kudos to Polaris and Snowflake for open-sourcing, kudos to Ali for open-sourcing Unity. It is a great thing for an industry. Let's figure out what customers want and then what's in there. There's no more black boxes.

>> From a product perspective, share your thoughts on open-source relative to developers standardizing around or formatting with data, because as data becomes generative, it's going to be in line in the coding cycle. So data is going to be part of coding where you're starting to see that now, not just in database. That part decoupling compute from data is going to enable some pretty cool things. What's the open-source angle? What's in it for the open-source participants? What's the ideal product scenario to evolve, if you had to guess that?

>> No, look, I think first you're going to get a ton of adoption when things are open. We've seen so many examples in our industry, Spark or Kafka and so on, and there are competing projects that didn't take off. You get overnight a ton of developers that are going to go commit and so on. You get a ton of free developers building, and then it still leaves vendors and opportunity to go and build the enterprise capabilities, kind of go innovate on top of that. But you are partnering with the open-source ecosystem. Where it goes wrong, is you try to do something that's against the customers, let the customers drive you, and as a product guy, you really want to go where the customers want. And the best way is go talk to a bunch of customers using the open-source and be like, what are the problems you're looking at and help solve that. So it becomes a lot easier as a product guy.

>> And what's your view on Iceberg and Delta Lake?

>> So I think they are two great table formats. I think we support both of them. Exciting to see how maybe both of them come together. I think the thing that I love is we are finally talking about a couple of formats, maybe one or two catalog formats as well. And we are going to spend... Actually, we've been spending a ton of time optimizing and moving data to Iceberg and Delta. So it's finally good to see open table formats, open metadata catalogs, and then we're going to be right there with our customers.

>> I'm curious, because I know the role of a chief product officer, there's a little bit of tension with marketing and some of the other teams. As you prioritize what's in your product roadmap, how do you approach that? It's a big challenge when things are moving as fast as they are right now.

>> No, I think it's a really good question. And often I think about quantitative and qualitative data, and if you go interview your best customers and maybe sort of thought leaders and you get a ton of data on how are customers using it, maybe some open source customers like we mentioned, and then you put that in front of... You sort of massage the data and then kind of how you prioritize becomes a boring exercise. But if you don't have the data, then it kind gets into a conflict. So Rachel and I, yeah, because in the data she's like, yeah, I think we should do that. Right? And then obviously there are strategic bets that you got to make and then do that with conviction.

>> You know your data is good if that conversation is boring. I really like that actually as a benchmark, I don't think I've ever had one of those conversations be boring. So I think that's a really. Well, obviously, your process is working, if you're figuring that. Do you feel like... I mean, you've been in data and analytics for a long time. It feels like the light is shining on your space a little bit more right now than normal. Would you agree? Is this a moment for you?

>> No, it's a great time for us. So if you think about, we've got many sort of tailwinds, AI. For you to build great models and great insight, the first thing you have to do is move data. We've invested in Iceberg and Delta, and this is important. We have 500 data sources that you can click a button and you get an Iceberg table or you get a Delta table. And that for a customer who's now thinking about that is less headache. It's hard to move that data when the schema changes, how do you think about updates? How do you think about governance? So we're just going to take all of those problems away for customers. So we feel like we've got two great trends, we've got great customers, and we've got over 200 customers using our Managed Data Lakes product. That's like, where are the pucks going?

>> What's the big drivers on the product-led growth, you guys are seeing, obviously you got a great product. Congratulations. I mean, having a simple product that's easy to use, that does a lot of high-performance capabilities is a winner. We've seen that. Is it the AI readiness or is it just data management in general's still growing? More data's coming in?

>> Yeah.

>> Or both. How would you rank the-

>> Actually, it's funny. We'd be surprised to find data's even growing on-prem. Data's going in the cloud. Data's going everywhere. We're just collecting more data, which is good for us, right? I think on the product-led growth, if a customer likes it and they're going to tell their LOB, they're going to tell their colleagues, it's really that. If the experience is gray, then the PLG sort of life cycle just go through. It's like a flywheel, we think. So really obsess about moving data, moving data correctly, moving to the right formats, and sort of tweaking all the knobs and you're like, "Hey, I just need to turn the button on. It works."

>> Everyone wants that easy start button.

>> Yeah.

>> All right. It's a very exciting time. Obviously you probably have a little... You mentioned there's an announcement coming tomorrow. I won't let you spoil it for us, but more broadly speaking, the next time we have your lovely self on our show, let's say at the summit next year, what do you hope to be able to say then that you can't say today?

>> I think I'd hope our customers would say if someone's moving data to Iceberg or Delta, they're like, "I'm just using ."

>> Perfect. Love the sound bite. We will be playing that back on our highlight reel when we have you back on the show. Thank you so much for taking the time to be here with us.

>> Absolutely.

>> Such an exciting.

>> Thanks for having me.

>> Hey, our pleasure, John. Great questions as always, and thank all of you for tuning into our two days of fantastic coverage here in our hometown of San Francisco, California. My name's Savannah Peterson here at Databricks Data and AI Summit. You're watching theCUBE, the leading source for enterprise tech news.