International User Groups Conference 2024 | Girish Pancha | CEO StreamSets (a Software AG company)

theCUBE.net

Clips
More from International User Groups Conference 2024

Girish Pancha

CEO

StreamSets (a Software AG Company)

play_circle_outline Exploring Data Integration Challenges: Girish Pancha on Cloud-Native Trends and Shifting from Transactional Data

play_circle_outline The role of AI in data integration and its potential in collaboration, productivity, and operations

play_circle_outline The upcoming migration of webMethods and StreamSets to IBM and the potential synergies and transformations it will bring

Info

Girish Pancha | CEO StreamSets (a Software AG company)

Girish Pancha

CEO StreamSets (a Software AG Company)

search

Paul Gillin

>> Welcome back. This is The Cube. I'm Paul Gillin. We're here at the Software AG International User Group Conference in Dublin Ireland, the calm before the storm here. Right now it's quiet behind us, but in a couple of hours, this place will be packed with customers of Software AG who represent, have a whole variety of technical disciplines. With us today before we get started, is Girish Pancha, who is Software AG's Chief Product Officer for the webMethods and the StreamSets product lines.

Girish Pancha

>> That's right.

Paul Gillin

>> And thanks for joining us today.

Girish Pancha

>> My pleasure. I'm excited to be here.

Paul Gillin

>> Now, you are an entrepreneur. You started StreamSets and you sold it to Software AG. The integration market is so crowded, there's so many different companies. What did you think you could do differently and do better?

Girish Pancha

>> Yeah. No, look, I've been in the integration space for a few decades. I was the Chief Product Officer of Informatica prior to that. So your question is very relevant, because when I started StreamSets, even my former colleagues wondered, what is there to do in data integration, let alone the broader integration space? And I didn't really think I was going to do StreamSets when I retired from Informatica. But I came to the realization that there were a number of very secular trends, disruptions that were happening in the data integration space. So first, was the re-architecting to a cloud native world. The second was the speeding up of these capabilities to keep in line with the business. So what we describe as batch to streaming. And the third was the shift away from, let's call it, transactional data, managing transactional data. In the old days when you talked about a customer, you'd think about, how much did they spend on you. So, you'd go to your application to figure out what that was. And what was happening was that we were moving to really trying to have a more holistic view of everything, customers, employees, our partners, etc. Where there was a lot more event and what we call interaction data that was around the edges, the term that was being used was big data in those days, obviously. And that kind of information was much less structured, much more fluid, always on. So these are kind of fundamental changes in the technology landscape that I figured the existing incumbents would've a hard time modernizing. Not to say it can't be done, but I thought there was at least kind of a time to market opportunity with StreamSets. And that's what .

Paul Gillin

>> A lot to unpack there. I mean, in terms of integration and bringing together all this structured and unstructured data, we're seeing the age of the data lake now, or the data lake house, the iceberg file format, some really transformative technology. How are you seeing that changing the integration challenges that your customers face?

Girish Pancha

>> Yeah, and that's kind of front and center. That's our sweet spot. So we started off dealing with, dare I say, the legacy technology that's Hadoop, and now most of our businesses around data lakes and data lake houses and iceberg, et cetera. At the highest level, what I would say is that it hurts or it makes it more difficult at design time because historically you had much better typed information in your source systems. And now you really don't. But I'd say equally importantly, when you talk about the runtime or the ops side of the world, the operations side of the world, because of the fluidity of those sources, things are much more prone to breaking and whatnot. And of course, on top of that, there's this expectation what DevOps did to applications. There's an expectation that we have an integration working at the speed of business. So all in all, you need the system to be continuous and the system to be able to rapidly deal with kind of unstructured, semi-structured unknown data. And so, that's the areas of focus that I had.

Paul Gillin

>> You also mentioned streaming. And that's an interesting topic because I mean, the research I've seen is that only about half of enterprises are using streaming in any capacity at all. Is that because the business doesn't demand it? Or are they missing out in many cases, on an opportunity?

Girish Pancha

>> I think it's the latter. I don't think... There's probably only maybe 10% of all the enterprises on earth that would say, "We don't need to do things that way." And of course, within each enterprise, we did hit buyers and users that said, "no, we don't care." Take equities for example. They make these bets. If it's not HFT, high-frequency trading, they're making these decisions and holding long. So, they don't need streaming every minute. They're happy to kind of look at it up periodically, whether it's daily, weekly, quarterly. There are different things there. But I would say everywhere we look generally within financial services, healthcare, all these verticals where we're very strong in, there is a need for streaming. The challenge, I think, is it's kind of two-fold. The existing technologies really did not handle it. So now you say, "Okay, well do I just kind of make do and maybe try to speed that up still to give me kind of a good enough solution? Or do I really start investing in a new technology in the same space to handle my streaming problems?" So the value of those use cases, I think, in both with verticals and to a certain extent, within the verticals, different companies, enterprises have different kind of DNAs in terms of, are they going to be early adopters, are they going to wait it out? And I think that's basically what it is. In fact, I would say, I could even say maybe I was a little too early to the game there. Because in 2015 when we launched the company, in fact the name is StreamSets. I focused on the set-based transmissions and the capabilities because people would say, "We don't need streaming." Well, a lot of them said, "We will bring you on for streaming in addition to the incumbents that are in the space." So, I think that's where we're at it. These things end up being, in my mind, a decade, multi-decade kind of transmission. And so, we're in the middle of that.

Paul Gillin

>> You certainly were ahead of your time, 2015. You have conceived this concept of... you're redefining iPaaS really, in your super iPaaS concept. What is that? How is your approach to integration different from everybody else's?

Girish Pancha

>> Yeah, look, so I mentioned at the beginning that we started off with this cloud-native approach to data integration. Probably about three, four years ago, we made first contact with Software AG. And of course, to certain folks, Software AG's got all sorts of technologies like mainframe technologies and whatnot. But I actually was very aware of webMethods as an app integration technology. I'd actually partnered with webMethods at my previous job to try to marry this idea of application integration and data integration into a single kind of combined, in this case, a multi-vendor, but a single offering. And we actually had been pretty successful with that in the 2000s. We called it a business activity platform to get these two together. So three, four years ago when I reconnected with Software AG, which had bought webMethods 15 years back, I got really excited, both about the higher level idea of marrying StreamSets with webMethods. But equally importantly, the fact that webMethods had made that shift to being cloud-native and being kind of a software service product too. I should say a hybrid, because obviously we're not leaving anything behind. So, there's a large install base that has webMethods deployed on-premise. But most of our new business now comes from the SaaS offering. And so, the SaaS offering is aware of the on-premise offering, and it works really well together. So when I learned this or when I saw this, I got really excited. I thought that this is one of those things that's been the holy grail. I think the business problem exists, the technology maturity wasn't there to really kind of put these two together, until everything got componentized and with no microservices and everything else. And now, I think the market maturity is starting to emerge. And it's going to take a little while because I give the example of just business productivity tools. There was a time when I first learned how to use a spreadsheet and people would talk about PowerPoint, I'd be like, "I don't know what that is and I don't care." Because I was a guy and that's all I cared about. And then suddenly, I couldn't buy Excel on its own. I had to buy Office or G Suite or whatever else. But along the way, I'm matured to realize that, "Oh, I can embed my Excel charts in my spreadsheets, et cetera." Sorry, my .

Paul Gillin

>> You're PowerPoint....

Girish Pancha

>> my PowerPoints. And once you start getting that, you suddenly realize you're opening up a broader set of use cases. Sometimes it may be I myself does all the work, but in other situations, it may be a collaboration. Somebody creates me that spreadsheet, I embed it in my PowerPoint, and off and running. And I don't really care about all the gory details of that spreadsheet. That's the analogy I think, that I would apply to the integration space, that there are components today. There's a lot of different personas that use these different components. But there is absolute business value in the long run for all of these things to be working together and ideally from a single vendor.

Paul Gillin

>> It's a wonderful metaphor you use. And of course, cloud-native constructs make this easier for these components to be integrated together. What about the role of AI. Your former employer, Informatica, is talking up a lot now about the value of AI and data integration. Is that a game-changer in data integration? Will your customers be able to use AI to make sense out of all this varying data and various formats and types?

Girish Pancha

>> Yeah. I definitely think it's a game changer. I think it's a bit of a quote, unquote, show me story. I mean, I could have said the same thing about OpenAI 10, 15 years ago, or let's not say 15, let's say seven, eight years ago when they're founded. And you would have to wait to see the end results before you actually said, "Okay, this is good enough and this is cool." I think the same thing is going to happen really in any vertical and definitely in the integration space, both data and app integration. And the reason being that the amount of data... So, let me break it down. So generative AI gives you the conversational interfaces. Great. But the domain-specific information which you need, which is easily available for generic conversations, the people are scouring the whole web for anything and everything there. But integration domain-specific information is not out there. It's actually locked into these products and technologies that we have out here. So I think he or she that figures that out, captures the most amount of integration data and then does the right things with it, is going to win because they're going to show the value. I'm going to be a bit of a contrarian. I don't like to knock, definitely not my former employee. And it's not to say we're not doing this, but really, we break down the AI application into three different buckets. The first is the collaboration with the lines of business. Now the emergence of business technologists that used to be called shadow IT and shadow IT basically state away from it. That was-

Paul Gillin

>> Now they're being integrated.

Girish Pancha

>> Now they're integrating, and now they're calling themselves-

Paul Gillin

>> champions....

Girish Pancha

>> business technology and enterprise guys are embracing them and they're embracing the enterprise, et cetera. So there's, I think, an opportunity to bring AI into that to help with that collaboration. So, that's one area. The second area, which is I think what everybody thinks about and assumes, is making those engineers productive, more productive. So what if instead of writing all this code, you can just magically say a few things and out pops enterprise-grade artifacts, data flows, and data pipelines or application flows and data pipelines, et cetera? So, that's a second bucket. And third bucket is around the operations. What if you could very quickly understand when anomalous things are happening, what the root cause analysis of that is? And have that in this conversational interface to get it out. So, I think those are the three buckets. And I feel everybody tends to talk about this middle bucket a lot. And I frankly, think that's the hardest. And I could even argue maybe it's the least valuable. I don't think we need to bring AI to get rid of one developer at this point in time. Maybe in the future we won't need any humans, but that's not where you want to optimize for right now and say, "We don't need developers. We can use AI." The other two areas, I think ripe because it's a new way of sharing information and thinking and developing things. And we'll either have to create a lot of tooling around it or maybe AI can help. And the operations, I mean, I just came from our customer advisory board, talk, not talk meeting this morning. And resoundingly, everybody has this operations problem. Because every time you do something complicated, you run it a hundred times, you run it a thousand times before you change whatever you did in the first place. And that's where when things go wrong, I was going to give you other statistics, like eight out of 10 people say this complex technology make everything very brittle. We have an operations problem. And I think if we can bring AI to that where we actually have a lot more data, 10 times as much runtime metadata as we have design time metadata. So if we can bring AI to that, I think that could be a cool game changer, at least in our space. And I would argue in many other spaces.

Paul Gillin

>> We'd be remiss if we didn't mention that things will be changing within the next few months. WebMethods and StreamSets are going to be migrating to IBM. You will become part of IBM, as will many of your people. What are you telling your customers right now about what this change means?

Girish Pancha

>> Yeah. So look, it's not just a product technology transfer, it's actual kind of full business unit transfer. So we have all the functions, go-to-market, sales, marketing, customer support, customer success, engineering, product, the whole shebangs heading over to IBM. So from our customer's perspective, they know that the only things that they may have to deal with differently, which they shouldn't have dealt at all, is some back office things. So I think that's like, let's call it the assuaging their fears, so to speak. So, that's a piece of it. But really what I would say is that we're all very excited about this because even as I talk about data and app integration, there are other things around, for example, event management, which we don't have, which IBM has. So I think actually, that we can bring more components into this super iPaaS framework. And of course, IBM's got a very, very deep technology stack in the data space, which we don't have. So, there's almost no overlap between StreamSets and what IBM has. So bringing all of those kinds of capabilities, data observability, data lineage, things that have existed forever, which we didn't focus on as a startup at StreamSets, bringing those and leveraging that. And of course, there's watsonx and the AI platform there. So you put that all together, I think there's technology synergies and excitement about what we can do together. There's excitement, I think, on their part because they've made a big bet on this. And we're actually going to be at scale and functioning the way we've been functioning. It is not just a tuck-in type acquisition at IBM. So, I think it can actually be transformational to the rest of the integration and automation stack at IBM. So generally, when I talk to our customers, say, "Obviously change always is uncomfortable. And you can always point to an acquisition by IBM that didn't go as well as you expected. But in this particular case, we should treat this as a specimen of one and go at it." So, that's what we're looking to do.

Paul Gillin

>> And my colleague, Dave Vellante, the Chief Analyst at theCUBE Research, very bullish on IBM's right now. Says that company is undergoing a transformation, thinks watsonx looks like a fantastic technology, and something that sounds very exciting for you.

Girish Pancha

>> Yeah. It's being run by a product leader for the first time in the history, I think, of the company. So, it reminds me a lot of what happened at Microsoft about 10 years ago. So, that's where I feel they're at right now.

Paul Gillin

>> That was exciting too. Girish Pancha, thank you very much for joining us, taking time out of your busy schedule at this busy conference.

Girish Pancha

>> My pleasure.

Paul Gillin

>> I'm Paul Gillin. This is theCUBE. We'll be right back.

Girish Pancha | CEO StreamSets (a Software AG company)

search

Paul Gillin

>> Welcome back. This is The Cube. I'm Paul Gillin. We're here at the Software AG International User Group Conference in Dublin Ireland, the calm before the storm here. Right now it's quiet behind us, but in a couple of hours, this place will be packed with customers of Software AG who represent, have a whole variety of technical disciplines. With us today before we get started, is Girish Pancha, who is Software AG's Chief Product Officer for the webMethods and the StreamSets product lines.

Girish Pancha

>> That's right.

Paul Gillin

>> And thanks for joining us today.

Girish Pancha

>> My pleasure. I'm excited to be here.

Paul Gillin

>> Now, you are an entrepreneur. You started StreamSets and you sold it to Software AG. The integration market is so crowded, there's so many different companies. What did you think you could do differently and do better?

Girish Pancha

>> Yeah. No, look, I've been in the integration space for a few decades. I was the Chief Product Officer of Informatica prior to that. So your question is very relevant, because when I started StreamSets, even my former colleagues wondered, what is there to do in data integration, let alone the broader integration space? And I didn't really think I was going to do StreamSets when I retired from Informatica. But I came to the realization that there were a number of very secular trends, disruptions that were happening in the data integration space. So first, was the re-architecting to a cloud native world. The second was the speeding up of these capabilities to keep in line with the business. So what we describe as batch to streaming. And the third was the shift away from, let's call it, transactional data, managing transactional data. In the old days when you talked about a customer, you'd think about, how much did they spend on you. So, you'd go to your application to figure out what that was. And what was happening was that we were moving to really trying to have a more holistic view of everything, customers, employees, our partners, etc. Where there was a lot more event and what we call interaction data that was around the edges, the term that was being used was big data in those days, obviously. And that kind of information was much less structured, much more fluid, always on. So these are kind of fundamental changes in the technology landscape that I figured the existing incumbents would've a hard time modernizing. Not to say it can't be done, but I thought there was at least kind of a time to market opportunity with StreamSets. And that's what .

Paul Gillin

>> A lot to unpack there. I mean, in terms of integration and bringing together all this structured and unstructured data, we're seeing the age of the data lake now, or the data lake house, the iceberg file format, some really transformative technology. How are you seeing that changing the integration challenges that your customers face?

Girish Pancha

>> Yeah, and that's kind of front and center. That's our sweet spot. So we started off dealing with, dare I say, the legacy technology that's Hadoop, and now most of our businesses around data lakes and data lake houses and iceberg, et cetera. At the highest level, what I would say is that it hurts or it makes it more difficult at design time because historically you had much better typed information in your source systems. And now you really don't. But I'd say equally importantly, when you talk about the runtime or the ops side of the world, the operations side of the world, because of the fluidity of those sources, things are much more prone to breaking and whatnot. And of course, on top of that, there's this expectation what DevOps did to applications. There's an expectation that we have an integration working at the speed of business. So all in all, you need the system to be continuous and the system to be able to rapidly deal with kind of unstructured, semi-structured unknown data. And so, that's the areas of focus that I had.

Paul Gillin

>> You also mentioned streaming. And that's an interesting topic because I mean, the research I've seen is that only about half of enterprises are using streaming in any capacity at all. Is that because the business doesn't demand it? Or are they missing out in many cases, on an opportunity?

Girish Pancha

>> I think it's the latter. I don't think... There's probably only maybe 10% of all the enterprises on earth that would say, "We don't need to do things that way." And of course, within each enterprise, we did hit buyers and users that said, "no, we don't care." Take equities for example. They make these bets. If it's not HFT, high-frequency trading, they're making these decisions and holding long. So, they don't need streaming every minute. They're happy to kind of look at it up periodically, whether it's daily, weekly, quarterly. There are different things there. But I would say everywhere we look generally within financial services, healthcare, all these verticals where we're very strong in, there is a need for streaming. The challenge, I think, is it's kind of two-fold. The existing technologies really did not handle it. So now you say, "Okay, well do I just kind of make do and maybe try to speed that up still to give me kind of a good enough solution? Or do I really start investing in a new technology in the same space to handle my streaming problems?" So the value of those use cases, I think, in both with verticals and to a certain extent, within the verticals, different companies, enterprises have different kind of DNAs in terms of, are they going to be early adopters, are they going to wait it out? And I think that's basically what it is. In fact, I would say, I could even say maybe I was a little too early to the game there. Because in 2015 when we launched the company, in fact the name is StreamSets. I focused on the set-based transmissions and the capabilities because people would say, "We don't need streaming." Well, a lot of them said, "We will bring you on for streaming in addition to the incumbents that are in the space." So, I think that's where we're at it. These things end up being, in my mind, a decade, multi-decade kind of transmission. And so, we're in the middle of that.

Paul Gillin

>> You certainly were ahead of your time, 2015. You have conceived this concept of... you're redefining iPaaS really, in your super iPaaS concept. What is that? How is your approach to integration different from everybody else's?

Girish Pancha

>> Yeah, look, so I mentioned at the beginning that we started off with this cloud-native approach to data integration. Probably about three, four years ago, we made first contact with Software AG. And of course, to certain folks, Software AG's got all sorts of technologies like mainframe technologies and whatnot. But I actually was very aware of webMethods as an app integration technology. I'd actually partnered with webMethods at my previous job to try to marry this idea of application integration and data integration into a single kind of combined, in this case, a multi-vendor, but a single offering. And we actually had been pretty successful with that in the 2000s. We called it a business activity platform to get these two together. So three, four years ago when I reconnected with Software AG, which had bought webMethods 15 years back, I got really excited, both about the higher level idea of marrying StreamSets with webMethods. But equally importantly, the fact that webMethods had made that shift to being cloud-native and being kind of a software service product too. I should say a hybrid, because obviously we're not leaving anything behind. So, there's a large install base that has webMethods deployed on-premise. But most of our new business now comes from the SaaS offering. And so, the SaaS offering is aware of the on-premise offering, and it works really well together. So when I learned this or when I saw this, I got really excited. I thought that this is one of those things that's been the holy grail. I think the business problem exists, the technology maturity wasn't there to really kind of put these two together, until everything got componentized and with no microservices and everything else. And now, I think the market maturity is starting to emerge. And it's going to take a little while because I give the example of just business productivity tools. There was a time when I first learned how to use a spreadsheet and people would talk about PowerPoint, I'd be like, "I don't know what that is and I don't care." Because I was a guy and that's all I cared about. And then suddenly, I couldn't buy Excel on its own. I had to buy Office or G Suite or whatever else. But along the way, I'm matured to realize that, "Oh, I can embed my Excel charts in my spreadsheets, et cetera." Sorry, my .

Paul Gillin

>> You're PowerPoint....

Girish Pancha

>> my PowerPoints. And once you start getting that, you suddenly realize you're opening up a broader set of use cases. Sometimes it may be I myself does all the work, but in other situations, it may be a collaboration. Somebody creates me that spreadsheet, I embed it in my PowerPoint, and off and running. And I don't really care about all the gory details of that spreadsheet. That's the analogy I think, that I would apply to the integration space, that there are components today. There's a lot of different personas that use these different components. But there is absolute business value in the long run for all of these things to be working together and ideally from a single vendor.

Paul Gillin

>> It's a wonderful metaphor you use. And of course, cloud-native constructs make this easier for these components to be integrated together. What about the role of AI. Your former employer, Informatica, is talking up a lot now about the value of AI and data integration. Is that a game-changer in data integration? Will your customers be able to use AI to make sense out of all this varying data and various formats and types?

Girish Pancha

>> Yeah. I definitely think it's a game changer. I think it's a bit of a quote, unquote, show me story. I mean, I could have said the same thing about OpenAI 10, 15 years ago, or let's not say 15, let's say seven, eight years ago when they're founded. And you would have to wait to see the end results before you actually said, "Okay, this is good enough and this is cool." I think the same thing is going to happen really in any vertical and definitely in the integration space, both data and app integration. And the reason being that the amount of data... So, let me break it down. So generative AI gives you the conversational interfaces. Great. But the domain-specific information which you need, which is easily available for generic conversations, the people are scouring the whole web for anything and everything there. But integration domain-specific information is not out there. It's actually locked into these products and technologies that we have out here. So I think he or she that figures that out, captures the most amount of integration data and then does the right things with it, is going to win because they're going to show the value. I'm going to be a bit of a contrarian. I don't like to knock, definitely not my former employee. And it's not to say we're not doing this, but really, we break down the AI application into three different buckets. The first is the collaboration with the lines of business. Now the emergence of business technologists that used to be called shadow IT and shadow IT basically state away from it. That was-

Paul Gillin

>> Now they're being integrated.

Girish Pancha

>> Now they're integrating, and now they're calling themselves-

Paul Gillin

>> champions....

Girish Pancha

>> business technology and enterprise guys are embracing them and they're embracing the enterprise, et cetera. So there's, I think, an opportunity to bring AI into that to help with that collaboration. So, that's one area. The second area, which is I think what everybody thinks about and assumes, is making those engineers productive, more productive. So what if instead of writing all this code, you can just magically say a few things and out pops enterprise-grade artifacts, data flows, and data pipelines or application flows and data pipelines, et cetera? So, that's a second bucket. And third bucket is around the operations. What if you could very quickly understand when anomalous things are happening, what the root cause analysis of that is? And have that in this conversational interface to get it out. So, I think those are the three buckets. And I feel everybody tends to talk about this middle bucket a lot. And I frankly, think that's the hardest. And I could even argue maybe it's the least valuable. I don't think we need to bring AI to get rid of one developer at this point in time. Maybe in the future we won't need any humans, but that's not where you want to optimize for right now and say, "We don't need developers. We can use AI." The other two areas, I think ripe because it's a new way of sharing information and thinking and developing things. And we'll either have to create a lot of tooling around it or maybe AI can help. And the operations, I mean, I just came from our customer advisory board, talk, not talk meeting this morning. And resoundingly, everybody has this operations problem. Because every time you do something complicated, you run it a hundred times, you run it a thousand times before you change whatever you did in the first place. And that's where when things go wrong, I was going to give you other statistics, like eight out of 10 people say this complex technology make everything very brittle. We have an operations problem. And I think if we can bring AI to that where we actually have a lot more data, 10 times as much runtime metadata as we have design time metadata. So if we can bring AI to that, I think that could be a cool game changer, at least in our space. And I would argue in many other spaces.

Paul Gillin

>> We'd be remiss if we didn't mention that things will be changing within the next few months. WebMethods and StreamSets are going to be migrating to IBM. You will become part of IBM, as will many of your people. What are you telling your customers right now about what this change means?

Girish Pancha

>> Yeah. So look, it's not just a product technology transfer, it's actual kind of full business unit transfer. So we have all the functions, go-to-market, sales, marketing, customer support, customer success, engineering, product, the whole shebangs heading over to IBM. So from our customer's perspective, they know that the only things that they may have to deal with differently, which they shouldn't have dealt at all, is some back office things. So I think that's like, let's call it the assuaging their fears, so to speak. So, that's a piece of it. But really what I would say is that we're all very excited about this because even as I talk about data and app integration, there are other things around, for example, event management, which we don't have, which IBM has. So I think actually, that we can bring more components into this super iPaaS framework. And of course, IBM's got a very, very deep technology stack in the data space, which we don't have. So, there's almost no overlap between StreamSets and what IBM has. So bringing all of those kinds of capabilities, data observability, data lineage, things that have existed forever, which we didn't focus on as a startup at StreamSets, bringing those and leveraging that. And of course, there's watsonx and the AI platform there. So you put that all together, I think there's technology synergies and excitement about what we can do together. There's excitement, I think, on their part because they've made a big bet on this. And we're actually going to be at scale and functioning the way we've been functioning. It is not just a tuck-in type acquisition at IBM. So, I think it can actually be transformational to the rest of the integration and automation stack at IBM. So generally, when I talk to our customers, say, "Obviously change always is uncomfortable. And you can always point to an acquisition by IBM that didn't go as well as you expected. But in this particular case, we should treat this as a specimen of one and go at it." So, that's what we're looking to do.

Paul Gillin

>> And my colleague, Dave Vellante, the Chief Analyst at theCUBE Research, very bullish on IBM's right now. Says that company is undergoing a transformation, thinks watsonx looks like a fantastic technology, and something that sounds very exciting for you.

Girish Pancha

>> Yeah. It's being run by a product leader for the first time in the history, I think, of the company. So, it reminds me a lot of what happened at Microsoft about 10 years ago. So, that's where I feel they're at right now.

Paul Gillin

>> That was exciting too. Girish Pancha, thank you very much for joining us, taking time out of your busy schedule at this busy conference.

Girish Pancha

>> My pleasure.

Paul Gillin

>> I'm Paul Gillin. This is theCUBE. We'll be right back.