Databricks Data + AI Summit 2024 | Jeff Veis & Anand Raman | Impetus Technologies

theCUBE.net

Clips
News
More from Databricks Data + AI Summit 2024

Jeff Veis

Chief Marketing Officer

Impetus Technologies

Anand Raman

Chief Revenue Officer

Impetus Technologies

Impetus Technologies sees continued momentum for AI as it speeds legacy data migration to Databricks

As a partner with Databricks Inc., Impetus Technologies Inc. has been active over the past year in legacy data migration of large-scale environments to the Databricks platform.It is important work for many organizations seeking to rapidly transition from legacy systems and realize the value of data analytics and generative AI.Jeff Veis and Anand Raman of Impetus Technologies talk with theCUBE about legacy data migration.“We’ve built IP that’s helped Databricks consume much, much faster,” said Anand Raman (pictured, right), chief revenue officer of Impetus Technologies

play_circle_outline Databricks' annual event, the Data + AI Summit

play_circle_outline Importance of a strong data foundation for gen AI

play_circle_outline Unlocking Automation and Migration Potential with Impetus Technologies' Flagship Tool, LeapLogic

play_circle_outline Driving Business Growth through Cost Optimization, Market Positioning, and Cloud Migration Strategies

Info

Jeff Veis & Anand Raman | Impetus Technologies

Jeff Veis

Chief Marketing Officer Impetus Technologies

Anand Raman

Chief Revenue Officer Impetus Technologies

search

>> Welcome back, everyone, to theCUBE's live coverage in San Francisco. I'm John Furrier, host of theCUBE, with Savannah Peterson. We're here for two days, wall-to-wall coverage, Databricks' annual event, the Data + AI Summit. We got a lot of coverage. Two great CUBE alumnis that have been around the block when it comes to data. This segment really could look at the historical perspective as well as the cutting edge new foundations that are being laid down, and get a glimpse into the future of AI and analytics. Jeff Veis CMO of Impetus Technologies, welcome back. Anand Raman, CRO, chief revenue officer Impetus Technologies. Guys, we've been on theCUBE having conversations from the big data days.

Anand Raman

>> Absolutely.

>> Hadoop, Spark Summit. Now this is like Spark Summit on steroids. It's called Data + AI Summit. Good to see you.

Anand Raman

>> Likewise. Likewise, John.

>> All right, so let's get into it. You guys have been riding this wave for a long time and we've been on this journey together. Big data, Hadoop. Again, a lot of things to work on, a lot of complex stack there. Didn't get the adoption income spark. Databricks takes that, jumps on the next lily pad, starts the data lake movement with more open technologies. Now, we got this generation of data now with generative AI where everyone is on a land grab for getting more capability. So, we're kind of in that, "Hey, we're finally here. Big data made it," moment.

Anand Raman

>> That's true. Absolutely.

>> We've seen this moving before, but now it's a completely different world.

Anand Raman

>> Yes.

>> What's your reaction to Databricks at this event?

Anand Raman

>> I think Databricks has done a fantastic job. We are coming to a place where really now the enterprises are able to make sense of the data. We internally at Impetus, we created a mission in 2017. We said, "How do you create powerful and intelligent enterprise?" And as we work through that, I think Databricks has been a very significant champion with us in helping enterprises become more intelligent and powerful.

>> Well, you guys are an elite tier partner of Databricks, which means what? Because they have a lot of partners. What is the elite tier?

Jeff Veis

>> We're the best.

>> Okay, so scope that for me. Give me an order measure. What does that mean? Level of client work? Is it scope of deals? What does that kind of partner level mean?

Anand Raman

>> It's a combination of things. What we've done for Databricks, we are one of the oldest partners of Databricks. We started with them, we've actually helped them help implement Databricks for many, many customers. Do some enterprise wide adoption of Databricks across companies. But more importantly, I think there are two things that have been really successful. One is we've built some IP that's helped Databricks consume much, much faster and enable Databricks to really make sure that the enterprises are making use of it. And for enterprises from whatever they've signed up with Databricks, be able to consume that quicker. So, we've been the fastest accelerator for Databricks consumption if you would, because we have a tool that helps enterprises migrate off of their legacy platform and transform all their 20 year, 30-year-old code into the new enterprise platforms like Databricks.

>> So, you guys have operating leverage, not just service.

Anand Raman

>> Absolutely.

>> You bring technology in. Okay, let's dig into that if you don't mind. Okay.

Anand Raman

>> Sure.

>> So, the trend clearly is, it's been happening for... It's a slow train wreck, legacy with data warehouses are moving into the cloud and these foundational models or driving everything. You can't just do it overnight.

Anand Raman

>> You can.

>> Take us through that work. What's involved? Are the data engineers doing it? Is it a full platform redo? What's involved in a migration or a switchover?

Anand Raman

>> Fantastic question and thanks for asking that because that's what our enterprises are really dealing with. And since you brought about a subject on history, a ten second, that's where Hadoop kind of failed. They built a new application but they couldn't help bridge that gap between legacy and this. And that's where we saw the open gap between the old technology and new, and which started our journey of building this product called LeapLogic. Like it says, it helps enterprises leap from the legacy world and take that logic as is and bring it and make it work on the new platforms. So, yes, it involves data engineering, but what it involves is a lot of R&D that's gone behind in building a product which gives up to 95% automation for enterprises.

>> So, as an example, can you bring that to the Unity catalog?

Anand Raman

>> Absolutely, we can bring it to the Unity catalog. And interestingly you bring Unity catalog to the picture. We've actually extended the data warehouse accelerator to even provide migration from Hive Metastore to Unity catalog. And what's that doing for Databricks is making enterprises ready for adoption of gen AI beyond all the governance and security and stuff. And again, getting rid of the legacy, moving to the world, new world and enabling them to get ready for gen AI.

>> So, it sounds like there's a lot of picks and shovels in the announcements this morning. We saw the mosaic tools are out there, so yeah, people are buildings a lot of picks and shovels, but what you're talking about foundationally, pour and nukes concrete, resetting the data foundation for the enterprise.

Jeff Veis

>> Well, that's a key thing and it's an interesting challenge because I know it's an overused term, but you need to have your data foundation really, really ready to move into the gen AI world. And that's easy to say, really, really hard to do, to be able to do that. The lift and shift shift to the cloud that organizations tried to do is not going to give you that data foundation. You need to be able to have it be governed, it needs to be able to be consumed and done in a trusted way. Only 30% of Databricks customers are on catalog today. So, there is a Grand Canyon that we have to get people through. And what's interesting about us is there's product companies, there's service companies, there's very few that do both IP and services. I've worked at both kinds of companies and the train kind of goes off the track quickly when you talk to a product company about, "Great, you have speeds and feeds, but how do I get there?" And what we're focused on is that join to bring the right automation technology that can do it in a smart way, but then bring the services around it so you can get the job done.

>> Well, it's interesting you bring that up because one, good business right now is having operating leverage with technology, but the service is what the customers need right now because they're dealing with a lot of legacy stuff that you just can't throw away. So, migration, I see that as a key thing. And plus all the other global service system integrators, they're all service-based. It's a service business.

Anand Raman

>> Yes.

>> Okay, so the question I have for you is what makes you guys different? Take us through the differentiators. Okay, I like what you're saying, sold me on it, but what about the other guys? What do they do? I hear the same thing. How are you different?

Anand Raman

>> I think fundamentally there are two things when you say difference. As Jeff pointed out, we are one of the only companies that brings deep maturity in product as well as services. I think there's one other concept, I'll let you talk about the three Es, but I think when you bring customer concept of innovation, reliability and zero disruption and you are able to prove that both in the short term and the long term, and that's the reason why customers like us. That's what differentiates us.

>> You go upfront, you lay that out upfront.

Anand Raman

>> We upfront lay that out. We have customers for whom we've built solutions and they've won patents on it. Enterprise customers who file for patents for the work that we've done and they won on it. So, when you bring that kind of work to a customer, they know what the value they're bringing because data is, every IT is turning into a software company today.

Jeff Veis

>> Yeah, actually one of these customers was good enough to put our engineer's name on the patent submission, and that's not usually what you get from a services company.

>> That's really good.

Jeff Veis

>> The three Es, since I'm the marketing guy.

>> Let's go.

Jeff Veis

>> Is the expertise, the engineering and the experience. And that may sound like something that everybody can speak to, but it's the join of those three that build up that we can look at the most demanding job, whether it's coming from a small company or a big established credit card company, you have to bring those three together. If you have just two or just one, you're usually not going to be terribly successful and you're going to go through a learning curve motion.

>> Just say the three Es again. Experience, engineering and-

Jeff Veis

>> Expertise.

>> Expertise. Okay. So, engineering right now I think critical. One of the things I've seen the past year and a half is engineering chops, minimum, table stakes.

Jeff Veis

>> Yes.

>> Because everything's being looked at from a data perspective and now with generative AI, all the plumbing and pipeline, it can be automated. So, agents are now going in place. So, now you can't get agents running if you don't get the data right. It's a systems redo. I mean ,it's a systems architecture, not just data.

Anand Raman

>> Absolutely.

>> It's a complete engineering assignment and that's where I think the experience comes in, to me. It's like, but let me rephrase this question. What experience is needed now that's new that wasn't around five years ago? If you could say learnings that are super important for generative AI that weren't around five years ago or so.

Anand Raman

>> I think if you look at generative AI, again there are three Es which kind of become very interesting. What is the experience that you give to an end user? What is the economy you bring to the end user? What is the efficiency you bring to the end user? So, if you combine those three Es of experience, efficiency and economy, how do you deliver that? Now that wasn't there five years ago. How do you interactively interact with data? Even five years ago, we were still talking about BIs and as you heard in the keynote today, it's not BI anymore, it's AI BI. So, you cannot live with static dashboards today. You have to work with the data interactively and that's what I think gen AI brings. But more importantly, as enterprises started to look at gen AI, what is now coming is, how do you really govern that? Because governance is going to become very, very important as you implement gen AI.

>> Governance has come up to the front lines. And that's a conversation that was only for a few conferences we go to.

Jeff Veis

>> But isn't that crazy though?

>> It's like the hottest thing on the planet right now because if you don't get it right, you can't let the data fly around.

Anand Raman

>> Exactly.

>> Because low latency data is going to be the next big thing.

Anand Raman

>> Yes.

>> How do you get low latency data? Just had the New York Stock Exchange ICE technology lead on earlier and we were talking about that. That's a huge factor. It's going to be huge.

Jeff Veis

>> I think you hit the nail on the head that you had these group, these categories that governance was like, the boring people did that, that you didn't want to talk to. You would hide when the governance people came because it meant pain and suffering. And then you would say the same thing with the visualization guys and the same thing with the data guys and the same thing with the apps guys and then there was the analytics guys. There were all these different groups that formed and usually a shiny light went on one or two of them. What's interesting here is there is no gap where you can just say, "You don't have to worry about that." It's the joy in bringing them together that this industry has spent the last 15 years kind of pulling apart to say, "You don't have to worry about that." And that is what is either I think going to separate the wheat from the chaff. The organizations that can understand how these things intertwined together are going to jump ahead. The ones that don't, they're going to be on the late night news because they recommended their competitor's product in their personalized ad campaign.

>> Well, they're going to be on the wrong side of history. They'll be out of business because they won't have a competitive advantage. I mean, to me, we were talking on theCUBE, I want to get your reaction to this because I agree with you. You hit a great point there. Right now the general consensus amongst most smart people in technology is like, I always ask the question, "What's this wave compared to other waves?" Because we've seen some waves, we've been on many waves. This one is big. And you'll see what people's reaction is. So, a few, I won't name names, they're pretty big deals, technologists, they don't want their names out there. They say, one said, 'It reminds me of email. Email and Office suite, productivity suite, because that changed everything." If you think about email, IT changed because of it. So, what you're getting at our earlier point about engineering a new foundation is like saying, "Hey, we just invented email. That means PCs and desktops. I'm going to have more productivity that you don't have to do other things on. Write down on-

Jeff Veis

>> I'm sorry, what's email? I haven't use that. What does that do?

>> I have so many inbox spam messages now. But that changed everything. But gen AI is going to do something similar where it's going to change the structural IT-

Jeff Veis

>> But let me talk on a serious note because I've been asked this before, why is this way different or is it different or is this just another hype cycle? In each hype cycle that I've seen and we don't have to getting to the cloud, getting to big data, there was inertia. There were companies that it was not in their best interest to have that new wave happen. So, you got hybrid computing, right? A little bit of cloud's okay, but you still need a data center, and et cetera. What is happening now is there's inertia with legacy hardware, but there are no inertia vendors that are saying, "Oh, no, you don't need AI."T There's not somebody sitting back and saying, this might be ill-advised." Everybody, from the smallest to the biggest, seems to be moving and you could argue AI washing, but everybody seems to be going, "Yep, this makes sense." And you didn't see that with big data. You had the big traditional database guys holding back. You didn't see that with the cloud. That's why cloud took 10, 15 years. You're not quite seeing that here with gen AI because I challenge you to find a company that goes, "Gen AI, just a fad. We'll get past this."

>> No, everyone's on board with this. It's obvious. It's just evolving.

Anand Raman

>> It is.

>> What's your reaction to the wave?

Anand Raman

>> No, I completely agree with Jeff. My sense is, as I said a few minutes ago, gen AI brings interactivity with data like never before. And I think that's what is going to be... And the multimodal aspect of it, bringing different things together. Your voice, your text, your images, your real time data, your batch data, your 10-year-old data and all the history and intelligence behind it. I think what that serves up, we were all impressed when the Siris and the Alexas came up and this is actually a complete next generation of the Siris and Alexas.

>> All those taglines in the big data days, "Put your data to work." Actually, it's in play right now.

Anand Raman

>> Yeah, absolutely.

>> All right, so let's go walk through what you guys are working on right now. You mentioned LogicLeap, what was it?

Anand Raman

>> LeapLogic.

>> LeapLogic. What other things you guys have in your arsenal, in your toolkit? Give me some of the features that you guys have, the key things for your customers.

Anand Raman

>> So, LeapLogic is one of the most important ones and it's our flagship tool where we are working with Databricks and all the cloud providers. It's one of the only tools, as I said, provides up to 95% of automation, zero disruption to the business and fail safe. We've done complex of Teradata migrations, complex to complex in record time that people thought they could never ever get rid of. Complex of ETLs, complex of analytical tools, everything now running in a Databricks environment. So, that's one. You brought up UC migration. So, we built accelerators to move from Hive Metastore to UC migration. Again, one of the only ones that can provide really 2x performance on the UC migration, and really creating the pathway for governance and gen AI.

>> Okay, so your relationship with Databricks is a good one, elite partner. What are some of the benefits to the customer? How does the pitch go? On the Teradata, you mentioned you guys, it's like 95%, no disruption, no fails. What's the pitch like? They go, "I don't believe you, prove it." Are they skeptical or they heard word of mouth that you guys got the product, they want to call you in? Take us through a day in the life of a sales call. Not sales call, but you know what I'm saying.

Anand Raman

>> No, I understand, and sometimes it is that it's too good to be true. And we have to show it to them and it's very simple. We have a simple web portal where somebody can put in their 100, 200, 500 lines of Teradata code and it'll spit out a Spark code or it'll spit out a DB SQL code right there. So, we-

>> What's the turnaround time? Instantly?

Anand Raman

>> Instantly.

>> What do you charged for that?

Anand Raman

>> From a proof of concept, that's free.

>> Okay. So, they go, -

Anand Raman

>> Have a web portal, we install it on AWS and Azure. Go and put your 500 lines of Teradata SQL, and-

>> And then what happens next? What's under the covers?

Jeff Veis

>> It will assess and tell you what can be done on an automated fashion, what is underutilized and perhaps you don't have to think about migrating it and what might need some customization. So, you'll get an assessment report that'll size that, that'll go through that. And that'll allow us to size the project, let you know what is involved and then allow you to be able to make a business decision.

>> You show the power tools and then they go, "Okay, great." Then you go through classic migration consulting. Okay, assessment with transformation. We got that operationalized.

Anand Raman

>> All of this is automated though.

Jeff Veis

>> Yeah, the tool does the assessment.

Anand Raman

>> All of this assessment is automated.

>> The tool does all assessment, service?

Anand Raman

>> We do the assessment. We were meeting with a CIO of a bank yesterday where we did the assessment and he came to meet us. He said, "This is the first time I know of my own application environment much better than I've ever known because your assessment tool told me what's being used, what's not being used, what's junk, what's technical debt and what really needs to get migrated."

>> So, what's the secret sauce? Years of experience that goes into training the tools and does it learn on its own? Give us a peek into the IP or the secret sauce?

Anand Raman

>> We've migrated about nearly billion plus lines of code today. So, the product has become intelligent in its migration and it understands, we don't do a line by line code migration, which many others in the competition do that. We understand patterns of how people write code and what patterns should focus where. And while we are understanding patterns, we also look at how these patterns have to run in the new environment. So, when we transform, the optimization and performance optimization of that pattern automatically happens. So, what the customer gets is if they're spending a million dollars on Teradata, when they come to the cloud, they'll either spend a million dollars or probably spend 900K.

>> So, you get good cost optimization.

Anand Raman

>> Absolutely.

>> So, that makes them feel good. They don't know they're going to overspend, so they get a good feel for what the impact is.

Anand Raman

>> And the return on investment is much better.

>> Yeah, TCO. Jeff, I mean, you've got to look at this and you've got to go, "This is kind of too good to be true." How do you sell through that? How do you market that? Just say -

Jeff Veis

>> Oh, you don't have to market, it's so good.

>> Word of mouth has got to be big.

Jeff Veis

>> Well, actually, I mean, what I hear a lot from the customers is we're admittedly a well-kept secret. And often I've worked for companies that had great marketing and maybe not so great product or product that wasn't quite ready for prime time. The situation with Impetus is the reverse. They've actually created a real, real differentiation and now it's about just telling the story. The market has moved into our sweet spot, so we're hopping in this wake now. By the way, data engineering is what we do. So, we're not spreading out and taking any SI project. If you want to install an HR system, there's lots of other vendors on the show floor we could point you to. Our passion, our focus is around data engineering and cloud.

>> You guys are like the fixer. You come in and you're going to clean up some stuff and reset the foundation because this wave is a disruptive enabler. Okay, it's going to enable but it's going to disrupt things. It's going to be old way and new way. Pretty soon we're going to start, as you pointed out, the side of the street, old way and new way. And it's becoming clear. In fact, the fog is kind of lifting a little bit. Say, okay, data warehouse, we all know what's going on there. I mean, Holly basically said it on his keynote, fragmented data estate. He's talking directly at what you're doing, right?

Anand Raman

>> Yeah.

Jeff Veis

>> That's straddling is the hardest thing to do. Every VC backed startup and high-flying company is pretty good about saying, "We got the next best thing. We got the next best to mousetrap.", The thing is, nobody's a greenfield. Everybody's starting with something no matter how small you are. And the trick is, how do you get from where you are to where you're trying to get to? And that's the spot we occupy.

>> And it must be reigning money. Chief revenue officer, clients are happy. They got a great tool to migrate, probably, which is a real pain point for them. I mean, this is probably a pain point that you help make go away.

Anand Raman

>> I think for us, the pain point that we are helping them go away is reliance on all the legacy systems. I think that's one of the most important thing. And the second pain point that as we meet with so many different new customers every day is how do I really make sure that my data strategy, our data infrastructure, is ready for the future?

>> Well, congratulations. You guys got a good opportunity. Put a quick plug in for the last 30 seconds we have left. What are you are you guys looking to do here? Databricks, beyond, what's next for the company? Give the pitch.

Anand Raman

>> No, I think the next for the company is definitely gen AI. There's a strong, very, very strong push on gen AI. With our experience with data engineering. I think we are setting ourselves up and setting our customers up for gen AI. I think as you've said, there's a lot more. It's a transformative journey. We are getting ready for it and we are ready to help our customers on it.

Jeff Veis

>> And I'll give a plug on top of that is one thing we have are these data and gen AI labs. For qualified clients, they're free. It's a couple of days getting together. It's looking at the strategy and design to be able to figure out which way you want to go from an objective standpoint. And that's something that we offer. We get a lot of uptake in that and it allows you to go from talk to actually put a kind of pen on paper, if you will, and get something that could be a blueprint for your future.

>> Well, Jeff, Anand, thanks for coming on theCUBE. Going to go back 2015 Spark Summit. That's nine years ago.

Anand Raman

>> That's right.

>> Nine years ago you were on. I mean, if you go back and look at that video, that was the beginning. This is all on a trajectory.

Jeff Veis

>> Yes.

>> A little bit higher evolution, but just we're feeling good right now.

Anand Raman

>> Absolutely, we are.

>> Congratulations.

Anand Raman

>> Thank you very much.

>> We're here live at San Francisco, theCUBE Databricks event. I'm John Furrier, your host with Savannah Peterson. We'll be right back after this short break.

Jeff Veis & Anand Raman | Impetus Technologies

search

>> Welcome back, everyone, to theCUBE's live coverage in San Francisco. I'm John Furrier, host of theCUBE, with Savannah Peterson. We're here for two days, wall-to-wall coverage, Databricks' annual event, the Data + AI Summit. We got a lot of coverage. Two great CUBE alumnis that have been around the block when it comes to data. This segment really could look at the historical perspective as well as the cutting edge new foundations that are being laid down, and get a glimpse into the future of AI and analytics. Jeff Veis CMO of Impetus Technologies, welcome back. Anand Raman, CRO, chief revenue officer Impetus Technologies. Guys, we've been on theCUBE having conversations from the big data days.

Anand Raman

>> Absolutely.

>> Hadoop, Spark Summit. Now this is like Spark Summit on steroids. It's called Data + AI Summit. Good to see you.

Anand Raman

>> Likewise. Likewise, John.

>> All right, so let's get into it. You guys have been riding this wave for a long time and we've been on this journey together. Big data, Hadoop. Again, a lot of things to work on, a lot of complex stack there. Didn't get the adoption income spark. Databricks takes that, jumps on the next lily pad, starts the data lake movement with more open technologies. Now, we got this generation of data now with generative AI where everyone is on a land grab for getting more capability. So, we're kind of in that, "Hey, we're finally here. Big data made it," moment.

Anand Raman

>> That's true. Absolutely.

>> We've seen this moving before, but now it's a completely different world.

Anand Raman

>> Yes.

>> What's your reaction to Databricks at this event?

Anand Raman

>> I think Databricks has done a fantastic job. We are coming to a place where really now the enterprises are able to make sense of the data. We internally at Impetus, we created a mission in 2017. We said, "How do you create powerful and intelligent enterprise?" And as we work through that, I think Databricks has been a very significant champion with us in helping enterprises become more intelligent and powerful.

>> Well, you guys are an elite tier partner of Databricks, which means what? Because they have a lot of partners. What is the elite tier?

Jeff Veis

>> We're the best.

>> Okay, so scope that for me. Give me an order measure. What does that mean? Level of client work? Is it scope of deals? What does that kind of partner level mean?

Anand Raman

>> It's a combination of things. What we've done for Databricks, we are one of the oldest partners of Databricks. We started with them, we've actually helped them help implement Databricks for many, many customers. Do some enterprise wide adoption of Databricks across companies. But more importantly, I think there are two things that have been really successful. One is we've built some IP that's helped Databricks consume much, much faster and enable Databricks to really make sure that the enterprises are making use of it. And for enterprises from whatever they've signed up with Databricks, be able to consume that quicker. So, we've been the fastest accelerator for Databricks consumption if you would, because we have a tool that helps enterprises migrate off of their legacy platform and transform all their 20 year, 30-year-old code into the new enterprise platforms like Databricks.

>> So, you guys have operating leverage, not just service.

Anand Raman

>> Absolutely.

>> You bring technology in. Okay, let's dig into that if you don't mind. Okay.

Anand Raman

>> Sure.

>> So, the trend clearly is, it's been happening for... It's a slow train wreck, legacy with data warehouses are moving into the cloud and these foundational models or driving everything. You can't just do it overnight.

Anand Raman

>> You can.

>> Take us through that work. What's involved? Are the data engineers doing it? Is it a full platform redo? What's involved in a migration or a switchover?

Anand Raman

>> Fantastic question and thanks for asking that because that's what our enterprises are really dealing with. And since you brought about a subject on history, a ten second, that's where Hadoop kind of failed. They built a new application but they couldn't help bridge that gap between legacy and this. And that's where we saw the open gap between the old technology and new, and which started our journey of building this product called LeapLogic. Like it says, it helps enterprises leap from the legacy world and take that logic as is and bring it and make it work on the new platforms. So, yes, it involves data engineering, but what it involves is a lot of R&D that's gone behind in building a product which gives up to 95% automation for enterprises.

>> So, as an example, can you bring that to the Unity catalog?

Anand Raman

>> Absolutely, we can bring it to the Unity catalog. And interestingly you bring Unity catalog to the picture. We've actually extended the data warehouse accelerator to even provide migration from Hive Metastore to Unity catalog. And what's that doing for Databricks is making enterprises ready for adoption of gen AI beyond all the governance and security and stuff. And again, getting rid of the legacy, moving to the world, new world and enabling them to get ready for gen AI.

>> So, it sounds like there's a lot of picks and shovels in the announcements this morning. We saw the mosaic tools are out there, so yeah, people are buildings a lot of picks and shovels, but what you're talking about foundationally, pour and nukes concrete, resetting the data foundation for the enterprise.

Jeff Veis

>> Well, that's a key thing and it's an interesting challenge because I know it's an overused term, but you need to have your data foundation really, really ready to move into the gen AI world. And that's easy to say, really, really hard to do, to be able to do that. The lift and shift shift to the cloud that organizations tried to do is not going to give you that data foundation. You need to be able to have it be governed, it needs to be able to be consumed and done in a trusted way. Only 30% of Databricks customers are on catalog today. So, there is a Grand Canyon that we have to get people through. And what's interesting about us is there's product companies, there's service companies, there's very few that do both IP and services. I've worked at both kinds of companies and the train kind of goes off the track quickly when you talk to a product company about, "Great, you have speeds and feeds, but how do I get there?" And what we're focused on is that join to bring the right automation technology that can do it in a smart way, but then bring the services around it so you can get the job done.

>> Well, it's interesting you bring that up because one, good business right now is having operating leverage with technology, but the service is what the customers need right now because they're dealing with a lot of legacy stuff that you just can't throw away. So, migration, I see that as a key thing. And plus all the other global service system integrators, they're all service-based. It's a service business.

Anand Raman

>> Yes.

>> Okay, so the question I have for you is what makes you guys different? Take us through the differentiators. Okay, I like what you're saying, sold me on it, but what about the other guys? What do they do? I hear the same thing. How are you different?

Anand Raman

>> I think fundamentally there are two things when you say difference. As Jeff pointed out, we are one of the only companies that brings deep maturity in product as well as services. I think there's one other concept, I'll let you talk about the three Es, but I think when you bring customer concept of innovation, reliability and zero disruption and you are able to prove that both in the short term and the long term, and that's the reason why customers like us. That's what differentiates us.

>> You go upfront, you lay that out upfront.

Anand Raman

>> We upfront lay that out. We have customers for whom we've built solutions and they've won patents on it. Enterprise customers who file for patents for the work that we've done and they won on it. So, when you bring that kind of work to a customer, they know what the value they're bringing because data is, every IT is turning into a software company today.

Jeff Veis

>> Yeah, actually one of these customers was good enough to put our engineer's name on the patent submission, and that's not usually what you get from a services company.

>> That's really good.

Jeff Veis

>> The three Es, since I'm the marketing guy.

>> Let's go.

Jeff Veis

>> Is the expertise, the engineering and the experience. And that may sound like something that everybody can speak to, but it's the join of those three that build up that we can look at the most demanding job, whether it's coming from a small company or a big established credit card company, you have to bring those three together. If you have just two or just one, you're usually not going to be terribly successful and you're going to go through a learning curve motion.

>> Just say the three Es again. Experience, engineering and-

Jeff Veis

>> Expertise.

>> Expertise. Okay. So, engineering right now I think critical. One of the things I've seen the past year and a half is engineering chops, minimum, table stakes.

Jeff Veis

>> Yes.

>> Because everything's being looked at from a data perspective and now with generative AI, all the plumbing and pipeline, it can be automated. So, agents are now going in place. So, now you can't get agents running if you don't get the data right. It's a systems redo. I mean ,it's a systems architecture, not just data.

Anand Raman

>> Absolutely.

>> It's a complete engineering assignment and that's where I think the experience comes in, to me. It's like, but let me rephrase this question. What experience is needed now that's new that wasn't around five years ago? If you could say learnings that are super important for generative AI that weren't around five years ago or so.

Anand Raman

>> I think if you look at generative AI, again there are three Es which kind of become very interesting. What is the experience that you give to an end user? What is the economy you bring to the end user? What is the efficiency you bring to the end user? So, if you combine those three Es of experience, efficiency and economy, how do you deliver that? Now that wasn't there five years ago. How do you interactively interact with data? Even five years ago, we were still talking about BIs and as you heard in the keynote today, it's not BI anymore, it's AI BI. So, you cannot live with static dashboards today. You have to work with the data interactively and that's what I think gen AI brings. But more importantly, as enterprises started to look at gen AI, what is now coming is, how do you really govern that? Because governance is going to become very, very important as you implement gen AI.

>> Governance has come up to the front lines. And that's a conversation that was only for a few conferences we go to.

Jeff Veis

>> But isn't that crazy though?

>> It's like the hottest thing on the planet right now because if you don't get it right, you can't let the data fly around.

Anand Raman

>> Exactly.

>> Because low latency data is going to be the next big thing.

Anand Raman

>> Yes.

>> How do you get low latency data? Just had the New York Stock Exchange ICE technology lead on earlier and we were talking about that. That's a huge factor. It's going to be huge.

Jeff Veis

>> I think you hit the nail on the head that you had these group, these categories that governance was like, the boring people did that, that you didn't want to talk to. You would hide when the governance people came because it meant pain and suffering. And then you would say the same thing with the visualization guys and the same thing with the data guys and the same thing with the apps guys and then there was the analytics guys. There were all these different groups that formed and usually a shiny light went on one or two of them. What's interesting here is there is no gap where you can just say, "You don't have to worry about that." It's the joy in bringing them together that this industry has spent the last 15 years kind of pulling apart to say, "You don't have to worry about that." And that is what is either I think going to separate the wheat from the chaff. The organizations that can understand how these things intertwined together are going to jump ahead. The ones that don't, they're going to be on the late night news because they recommended their competitor's product in their personalized ad campaign.

>> Well, they're going to be on the wrong side of history. They'll be out of business because they won't have a competitive advantage. I mean, to me, we were talking on theCUBE, I want to get your reaction to this because I agree with you. You hit a great point there. Right now the general consensus amongst most smart people in technology is like, I always ask the question, "What's this wave compared to other waves?" Because we've seen some waves, we've been on many waves. This one is big. And you'll see what people's reaction is. So, a few, I won't name names, they're pretty big deals, technologists, they don't want their names out there. They say, one said, 'It reminds me of email. Email and Office suite, productivity suite, because that changed everything." If you think about email, IT changed because of it. So, what you're getting at our earlier point about engineering a new foundation is like saying, "Hey, we just invented email. That means PCs and desktops. I'm going to have more productivity that you don't have to do other things on. Write down on-

Jeff Veis

>> I'm sorry, what's email? I haven't use that. What does that do?

>> I have so many inbox spam messages now. But that changed everything. But gen AI is going to do something similar where it's going to change the structural IT-

Jeff Veis

>> But let me talk on a serious note because I've been asked this before, why is this way different or is it different or is this just another hype cycle? In each hype cycle that I've seen and we don't have to getting to the cloud, getting to big data, there was inertia. There were companies that it was not in their best interest to have that new wave happen. So, you got hybrid computing, right? A little bit of cloud's okay, but you still need a data center, and et cetera. What is happening now is there's inertia with legacy hardware, but there are no inertia vendors that are saying, "Oh, no, you don't need AI."T There's not somebody sitting back and saying, this might be ill-advised." Everybody, from the smallest to the biggest, seems to be moving and you could argue AI washing, but everybody seems to be going, "Yep, this makes sense." And you didn't see that with big data. You had the big traditional database guys holding back. You didn't see that with the cloud. That's why cloud took 10, 15 years. You're not quite seeing that here with gen AI because I challenge you to find a company that goes, "Gen AI, just a fad. We'll get past this."

>> No, everyone's on board with this. It's obvious. It's just evolving.

Anand Raman

>> It is.

>> What's your reaction to the wave?

Anand Raman

>> No, I completely agree with Jeff. My sense is, as I said a few minutes ago, gen AI brings interactivity with data like never before. And I think that's what is going to be... And the multimodal aspect of it, bringing different things together. Your voice, your text, your images, your real time data, your batch data, your 10-year-old data and all the history and intelligence behind it. I think what that serves up, we were all impressed when the Siris and the Alexas came up and this is actually a complete next generation of the Siris and Alexas.

>> All those taglines in the big data days, "Put your data to work." Actually, it's in play right now.

Anand Raman

>> Yeah, absolutely.

>> All right, so let's go walk through what you guys are working on right now. You mentioned LogicLeap, what was it?

Anand Raman

>> LeapLogic.

>> LeapLogic. What other things you guys have in your arsenal, in your toolkit? Give me some of the features that you guys have, the key things for your customers.

Anand Raman

>> So, LeapLogic is one of the most important ones and it's our flagship tool where we are working with Databricks and all the cloud providers. It's one of the only tools, as I said, provides up to 95% of automation, zero disruption to the business and fail safe. We've done complex of Teradata migrations, complex to complex in record time that people thought they could never ever get rid of. Complex of ETLs, complex of analytical tools, everything now running in a Databricks environment. So, that's one. You brought up UC migration. So, we built accelerators to move from Hive Metastore to UC migration. Again, one of the only ones that can provide really 2x performance on the UC migration, and really creating the pathway for governance and gen AI.

>> Okay, so your relationship with Databricks is a good one, elite partner. What are some of the benefits to the customer? How does the pitch go? On the Teradata, you mentioned you guys, it's like 95%, no disruption, no fails. What's the pitch like? They go, "I don't believe you, prove it." Are they skeptical or they heard word of mouth that you guys got the product, they want to call you in? Take us through a day in the life of a sales call. Not sales call, but you know what I'm saying.

Anand Raman

>> No, I understand, and sometimes it is that it's too good to be true. And we have to show it to them and it's very simple. We have a simple web portal where somebody can put in their 100, 200, 500 lines of Teradata code and it'll spit out a Spark code or it'll spit out a DB SQL code right there. So, we-

>> What's the turnaround time? Instantly?

Anand Raman

>> Instantly.

>> What do you charged for that?

Anand Raman

>> From a proof of concept, that's free.

>> Okay. So, they go, -

Anand Raman

>> Have a web portal, we install it on AWS and Azure. Go and put your 500 lines of Teradata SQL, and-

>> And then what happens next? What's under the covers?

Jeff Veis

>> It will assess and tell you what can be done on an automated fashion, what is underutilized and perhaps you don't have to think about migrating it and what might need some customization. So, you'll get an assessment report that'll size that, that'll go through that. And that'll allow us to size the project, let you know what is involved and then allow you to be able to make a business decision.

>> You show the power tools and then they go, "Okay, great." Then you go through classic migration consulting. Okay, assessment with transformation. We got that operationalized.

Anand Raman

>> All of this is automated though.

Jeff Veis

>> Yeah, the tool does the assessment.

Anand Raman

>> All of this assessment is automated.

>> The tool does all assessment, service?

Anand Raman

>> We do the assessment. We were meeting with a CIO of a bank yesterday where we did the assessment and he came to meet us. He said, "This is the first time I know of my own application environment much better than I've ever known because your assessment tool told me what's being used, what's not being used, what's junk, what's technical debt and what really needs to get migrated."

>> So, what's the secret sauce? Years of experience that goes into training the tools and does it learn on its own? Give us a peek into the IP or the secret sauce?

Anand Raman

>> We've migrated about nearly billion plus lines of code today. So, the product has become intelligent in its migration and it understands, we don't do a line by line code migration, which many others in the competition do that. We understand patterns of how people write code and what patterns should focus where. And while we are understanding patterns, we also look at how these patterns have to run in the new environment. So, when we transform, the optimization and performance optimization of that pattern automatically happens. So, what the customer gets is if they're spending a million dollars on Teradata, when they come to the cloud, they'll either spend a million dollars or probably spend 900K.

>> So, you get good cost optimization.

Anand Raman

>> Absolutely.

>> So, that makes them feel good. They don't know they're going to overspend, so they get a good feel for what the impact is.

Anand Raman

>> And the return on investment is much better.

>> Yeah, TCO. Jeff, I mean, you've got to look at this and you've got to go, "This is kind of too good to be true." How do you sell through that? How do you market that? Just say -

Jeff Veis

>> Oh, you don't have to market, it's so good.

>> Word of mouth has got to be big.

Jeff Veis

>> Well, actually, I mean, what I hear a lot from the customers is we're admittedly a well-kept secret. And often I've worked for companies that had great marketing and maybe not so great product or product that wasn't quite ready for prime time. The situation with Impetus is the reverse. They've actually created a real, real differentiation and now it's about just telling the story. The market has moved into our sweet spot, so we're hopping in this wake now. By the way, data engineering is what we do. So, we're not spreading out and taking any SI project. If you want to install an HR system, there's lots of other vendors on the show floor we could point you to. Our passion, our focus is around data engineering and cloud.

>> You guys are like the fixer. You come in and you're going to clean up some stuff and reset the foundation because this wave is a disruptive enabler. Okay, it's going to enable but it's going to disrupt things. It's going to be old way and new way. Pretty soon we're going to start, as you pointed out, the side of the street, old way and new way. And it's becoming clear. In fact, the fog is kind of lifting a little bit. Say, okay, data warehouse, we all know what's going on there. I mean, Holly basically said it on his keynote, fragmented data estate. He's talking directly at what you're doing, right?

Anand Raman

>> Yeah.

Jeff Veis

>> That's straddling is the hardest thing to do. Every VC backed startup and high-flying company is pretty good about saying, "We got the next best thing. We got the next best to mousetrap.", The thing is, nobody's a greenfield. Everybody's starting with something no matter how small you are. And the trick is, how do you get from where you are to where you're trying to get to? And that's the spot we occupy.

>> And it must be reigning money. Chief revenue officer, clients are happy. They got a great tool to migrate, probably, which is a real pain point for them. I mean, this is probably a pain point that you help make go away.

Anand Raman

>> I think for us, the pain point that we are helping them go away is reliance on all the legacy systems. I think that's one of the most important thing. And the second pain point that as we meet with so many different new customers every day is how do I really make sure that my data strategy, our data infrastructure, is ready for the future?

>> Well, congratulations. You guys got a good opportunity. Put a quick plug in for the last 30 seconds we have left. What are you are you guys looking to do here? Databricks, beyond, what's next for the company? Give the pitch.

Anand Raman

>> No, I think the next for the company is definitely gen AI. There's a strong, very, very strong push on gen AI. With our experience with data engineering. I think we are setting ourselves up and setting our customers up for gen AI. I think as you've said, there's a lot more. It's a transformative journey. We are getting ready for it and we are ready to help our customers on it.

Jeff Veis

>> And I'll give a plug on top of that is one thing we have are these data and gen AI labs. For qualified clients, they're free. It's a couple of days getting together. It's looking at the strategy and design to be able to figure out which way you want to go from an objective standpoint. And that's something that we offer. We get a lot of uptake in that and it allows you to go from talk to actually put a kind of pen on paper, if you will, and get something that could be a blueprint for your future.

>> Well, Jeff, Anand, thanks for coming on theCUBE. Going to go back 2015 Spark Summit. That's nine years ago.

Anand Raman

>> That's right.

>> Nine years ago you were on. I mean, if you go back and look at that video, that was the beginning. This is all on a trajectory.

Jeff Veis

>> Yes.

>> A little bit higher evolution, but just we're feeling good right now.

Anand Raman

>> Absolutely, we are.

>> Congratulations.

Anand Raman

>> Thank you very much.

>> We're here live at San Francisco, theCUBE Databricks event. I'm John Furrier, your host with Savannah Peterson. We'll be right back after this short break.