Exploring AI Integration in Banking and Data Management at the Databricks DATA+AI Summit
Christian Nelissen, chief data and analytics officer at National Australia Bank, and Ed Lenta, Senior Vice President and General Manager at Databricks, join John Furrier of theCUBE at the Databricks DATA+AI Summit in San Francisco. This session explores evolving data strategies in the banking sector, with a particular focus on AI integration and management advancement.
Nelissen discusses the transformative journey National Australia Bank embarked on three years ago with Databricks. Believing in the platform's potential, the bank commits to Databricks, resulting in a productive partnership driven by a shared vision. Lenta and Furrier emphasize how this collaboration exemplifies the bank's commitment to innovation within its data infrastructure. This commitment receives support from executives and researchers from theCUBE.
Key takeaways include the significance of a clear mission to facilitate data management and AI integration, as highlighted by Nelissen. The discussion further addresses the challenges and triumphs encountered throughout the banking transformation process, showcasing the tandem efforts of National Australia Bank and Databricks as facilitated by unity and strategic foresight, according to Lenta.
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
Data + AI Summit 2025. If you don’t think you received an email check your
spam folder.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open this link to automatically sign into the site.
Register For Data + AI Summit 2025
Please fill out the information below. You will recieve an email with a verification link confirming your registration. Click the link to automatically sign into the site.
You’re almost there!
We just sent you a verification email. Please click the verification button in the email. Once your email address is verified, you will have full access to all event content for Data + AI Summit 2025.
I want my badge and interests to be visible to all attendees.
Checking this box will display your presense on the attendees list, view your profile and allow other attendees to contact you via 1-1 chat. Read the Privacy Policy. At any time, you can choose to disable this preference.
Select your Interests!
add
Upload your photo
Uploading..
OR
Connect via Twitter
Connect via Linkedin
EDIT PASSWORD
Share
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
Data + AI Summit 2025. If you don’t think you received an email check your
spam folder.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open this link to automatically sign into the site.
Sign in to gain access to Data + AI Summit 2025
Please sign in with LinkedIn to continue to Data + AI Summit 2025. Signing in with LinkedIn ensures a professional environment.
Are you sure you want to remove access rights for this user?
Details
Manage Access
email address
Community Invitation
Christian Nelissen, National Australia Bank & Ed Lenta, Databricks
Exploring AI Integration in Banking and Data Management at the Databricks DATA+AI Summit
Christian Nelissen, chief data and analytics officer at National Australia Bank, and Ed Lenta, Senior Vice President and General Manager at Databricks, join John Furrier of theCUBE at the Databricks DATA+AI Summit in San Francisco. This session explores evolving data strategies in the banking sector, with a particular focus on AI integration and management advancement.
Nelissen discusses the transformative journey National Australia Bank embarked on three years ago with Databricks. Believing in the platform's potential, the bank commits to Databricks, resulting in a productive partnership driven by a shared vision. Lenta and Furrier emphasize how this collaboration exemplifies the bank's commitment to innovation within its data infrastructure. This commitment receives support from executives and researchers from theCUBE.
Key takeaways include the significance of a clear mission to facilitate data management and AI integration, as highlighted by Nelissen. The discussion further addresses the challenges and triumphs encountered throughout the banking transformation process, showcasing the tandem efforts of National Australia Bank and Databricks as facilitated by unity and strategic foresight, according to Lenta.
Christian Nelissen, National Australia Bank & Ed Lenta, Databricks
Christian Nelissen
Chief Data and Analytics OfficerNational Australia Bank
Ed Lenta
SVP and GMDatabricks
Christian Nelissen, chief data and analytics officer of the National Australia Bank, and Ed Lenta, senior vice president and general manager, APJ, at Databricks Inc., join theCUBE’s John Furrier at the Databricks Data + AI Summit 2025. The discussion examines how strategic partnerships and AI innovation are transforming data environments in the financial services sector.
Nelissen shares how the National Australia Bank’s three-year partnership with Databricks has enabled a confident shift away from legacy systems. Lenta highlights how shared goals and...Read more
Christian Nelissen, National Australia Bank & Ed Lenta, Databricks
search
>> Welcome back everyone to the live coverage here at theCUBE. We're here in San Francisco for Databricks Data + AI Summit. My name is John Furrier. I'm the host of theCUBE, wall-to-wall coverage, all the top people, the senior engineers, the founders, scientists, bringing in the AI error and the AI-powered banking, big keynote up on stage from a big bank here talking about leaning into AI. Of course, data's like electricity and if you're a bank you got a lot of it. We have two great guys, Christian, Chief Data and Analyst Officer in National Australia Bank. Thanks for coming on. I appreciate it.>> Thank you.>> And SVP and GM for Databricks and APJ.>> Great to be on. Thank you so much, John.>> You own the territory there. You got that whole set of continents out there.>> It is certainly an enormous space to be doing businesses. That's true.>> I got to say first, congratulations on, again, the platform evolution on Databricks, continuing to advance the data infrastructure, the apps, and obviously the agent bricks looking very strong. Christian, love to get your perspective. Obviously a lot of customers love Databricks because last year they had bought Tabular and Delta and Iceberg became the de facto standard. This year, so much has happened. As analytics and engineering of AI come together, what's your take as a customer for Databricks? What's your view on where they are now, their evolution of their platform?>> We're super excited to be partners with them, obviously. We've been on a journey with them. We made the decision to go with them about three years ago and at that point, not everything that they now have had been delivered, but we had a huge amount of confidence that it was going to get delivered. Once you met the senior guys, they gave us confidence that these were super smart guys with a real clear vision of the roadmap and the product was already really good, but just had so much more potential that we took a bit of a punt, but we feel really good about the decision that we've made.>> And say when you jump out of the plane, the parachute, I hope it opens, right?>> Yeah, yeah. I had a conversation with Ali where you'll remember this, I think early on we had a conversation with Ali that where I said, "Just so you understand, we're all in on Databricks. There's not going to be anybody else. We're all in on you." But that means it has to work both ways because it's a decent-sized risk for us, but we're happy with the way it's worked out.>> Hey, talk about the culture of Databricks because we've been covering Databricks since before they even got their first round of funding out of Cal Berkeley and then watch the progression. Just the smart brain power, the vision of open, the vision of making things simpler has always been a north star for Databricks. You got to be pretty happy to see that, but all in bet puts a lot of pressure on the Databricks team and the engineers and they like to get their use cases because they're learning from them. Share that cultural vibe there. I used the word vibe. Vibe coatings last year, now it's engineering this year. Engineer, good answer there.>> We love to be working with the National Australia Bank and it's true that all in and all of the seriousness that comes with that puts a lot of pressure on you. But without that clarity of purpose, John, you don't get things done. And so when customers say to me, "This is a big journey that we're embarking on, how are we successful in the AI space," the key ingredient is that, is clarity of purpose. Do you have the will and the fortitude to burn the boats, to leave the way that you did it behind you and to move forward on this journey? And then the other thing I'd say is we're constantly challenged by the National Australia Bank and we constantly challenge them back and it's just an amazing partnership. When you can work together for three or four years with enormous clarity of purpose to get something done, you can achieve incredible things.>> Christian, talk about the process, change management is the hardest thing to do. You have to burn the boats in some cases and say, "There's only one path but forward. We're going to survive.">> 100%. When I joined, we had two legacy data warehouses plus one that we built ourselves in the cloud. We're a very cloud native bank. We think we're the second-highest cloud adoption bank in the world. So 83% of all of our workloads are on the cloud, but we'd already had one go at building a cloud base and you come in and you go, "Okay, well, you've just invested all this on this, this cloud-based thing. We're going to park all of that and we're going to change direction." It's a huge, huge shift. Had huge support from the CEO and from my other two bosses. Most of the CIO and the COO made a big difference, but it was hard work. You had to grind it out. We shut down Teradata, 26-year-old estate. We shut that down in the first two years, moved everything into Databricks. It was a huge achievement. Every one of those things, every step was a battle. And I have to tell you, we were a month out from turning it off and we'd done all the comms and somebody from somewhere emailed us and said, "I don't think I've got a use case on that and you can't turn it off." And we're a month out of all this and every day a battle. The team did an amazing job to turn it off, but 26-year-old warehouse migrated.>> Beat it with a stick.>> 100%.>> In the grind. There's always that one straggler. So this is migration pain. If you look at today's announcements, migration has definitely stepped up its game in terms of some of the tooling. Have you been part of some of the projects? Because a lot of their new stuff has been tests rolled with customers. Which have you been involved in?>> I often say to Databricks, "I'd love to live in a world where you are no longer on my critical path." So it will tell you the number of conversations we've had where when's this coming, we're waiting for it, we said April. It's a good healthy tension.>> Carriers are sticking back in the Databricks. Let's go.>> As Ed says, it's a partnership. You understand that not everything comes when it needs to come, but it often does come and when it comes, it's always really, really strong and really powerful.>> Yeah, Jonathan Frankle was just on, he's the chief science, came from Mosaic. He and I were just talking. He was saying he's fascinated with reading old computer science manuals from the '70s. I'm like, "Okay, first of all, that's wrong generation, that's punch cards." But the theory at that time was programming, no matter how much you coded, it didn't matter because you couldn't go very fast. But once the speed came, his point was that the engineers didn't know what to do because they've never seen speed before. So we're in an era now where you got all this new capability, it's an unknown that we're moving into, so you got to move fast, but you got to understand what's going on. It's a whole nother world. So we're seeing that same kind of environment.>> 100%. But also, one of the things we're seeing, too, is that the great thing about Databricks, particularly as all the new functionality comes on, is we can move away from coding to configuration. And that makes a huge difference for us in terms of the amount of workload and the standardization. That's been a really important part of the process as well. You also mentioned, though, at the beginning of our discussion, this notion of data like electricity. And so that's what Christian made completely clear to us up front. He said, "No, I want data to flow like electricity through the National Australia Bank and then to be able to have the business extract value from it at any point in that." And so the mission was clear, how do we get everything real time? How do we get everything moving at speed through the bank and how do we remove anything that's in the way? And so that just made it a lot easier to say, "Okay, well, we now have declarative frameworks for ingestion and pipelines. We now have ways to connect to source systems." It just made it a lot simpler.>> Great point. Christian, electricity, and he mentioned the word value extraction. Sometimes we get caught into the value creation side of things, but if you don't have the extraction, you're in that business. You live to create an environment so that value can create it, you take their creation, you create value and you extract it. How do you think about the value extraction from your perspective?>> The first thing I'd say in what I do is because we're trying to solve a problem that the whole bank has and it's really easy to get hung up on the particular use case. And as soon as I find if I get hung up on a particular use case, the whole organization starts to say, "Okay, well, what would it take to deliver that use case?" And you keep having to bring them back out and say, "No, I'm trying to create a platform that solves for all use cases."
And I think if you can hold the line on that long enough, you get to a point where you start to really deliver value. And the interesting thing is we've got now so much data on Databricks that people who come to us to do new things... I've got a great example. We have a use case, it needs 43 sources worth of data, which is quite a lot. So different core banking systems need to combine. We've got 40 of those sources already on Databricks that we can reuse. We need three more. So the bill for them to do that work has gone from tens of millions of dollars to->> A fraction of that.>> That's right.>> So this is a really big point about platforms and leverage and reuse, operating leverage and reuse. The old days of silos, that's my budget, well, stay in your lane. Hate that word, stay in your lane. When I hear that, I'm like, "Whoa, whoa. This is fighting words. Stay in your lane. There's no lanes." But if you have that foundation, that lake now, lake base and these other things, you're enabling a foundational service>> Data like electricity.>> Yeah.>> That's the thing, the reason we came up with, at least my version of it was, when we came up with it, it was you don't want to spend any time thinking about. Like electricity, you just think it's going to be there and you plug it in and away you go, 100%. It's the most valuable thing I think you could do is to make data disappear as a problem.>> All right, so talk about data governance. You're in the banking business, so you live this.>> Yeah.>> We're hearing a lot here about governance built in. A lot of this AI's are going to take care of that. What are you seeing on the whole governance compliance regulation that being agents-driven, human judges?>> We love Unity Catalog. It's our superpower for anytime somebody says to us how you control Unity Catalog, it gives us column level, row level, access control to our sensitive personal information, our personal information, our private information. We have geographical restrictions on some country. Unless you operate in a particular country, you're not allowed to see data. All of that we control through Unity Catalog. And the arguments about, "Oh, we've got all this legacy estate, the heritage estate that we need to fix," one of the answers to if you want to be secure and well governed, move it onto the platform because we can deliver high quality control data.>> You're a blueprint.>> 100%.>> You're a blueprint for the legacy.>> Yeah, 100%.>> Which is night and day, really, because three years ago when Christian and I started working together, the biggest complaint that Christian had about the current data estate was one of trust, that the businesses losing trust in the quality and the heritage and the lineage of the data. How do we bring that back? And so to hear this now is really music to my ears. And another interesting thing, John, is governance, security, access and so on has been a key piece of the work we've done. But now we're working together co-innovating to say, "Actually, can we make Databricks the platform for security on the bank? Can we make Databricks the way that the bank ingests data from all the machines and various capabilities and become that data platform?">> The theme here to show is evaluation, evaluate the models, evaluate your security posture. You don't have to kill innovation to maintain a posture where Jamie Dimon from JPMorgan Chase say that. If you don't lean into it, my words, you'll be up the creek without a paddle. So he didn't actually say that. That's my interpretation.>> He did an amazing job.>> He basically said that. All right, so now the use cases that you're enabling, are there some obvious things that you're seeing? You guys work from the patterns of the customers you have, starting to see usage patterns. You mentioned some of the examples. What are the top use cases you guys are knocking down now and what are the ones you're going after?>> You see across the region, why don't you->> Yeah, so we work with many banking customers across the region. Security is a great one. So how do we secure the firm? How do we make our company more secure by gathering huge amounts of events and huge amounts of data? In financial services that is huge. But then that morphs to things like financial crime, everything around your customer is another big one. And today, John, it remains true across every vertical that the single most widely used use case is the customer data platform. How do I know my customers better and how do I provide them with more timely offers that are more specific to them in a way that's going to drive top-line revenue?>> And personalization is killer ad for AI.>> 100%. We use it->> You feel like I have a personal relationship with the bank, they know me.>> 100%.>> Not you know the customer, but from a regulatory perspective but more from a service for them.>> Of course.>> And why wouldn't they?>> Of course. And I would imagine that the marketing team at the National Australia Bank is excited about that.>> Yeah, 100%. We use it to feed our personalization engine and it makes a huge difference because the more information, the more data you can get in there, the more targeted you can be about when you need to talk to a customer what to talk to them about.>> Ed and Christian, I love the story here. It's a transformation story. Talk about the origination, first meeting, Databricks guys are coming in.>> I wish I could remember.>> I got to hear the pitch. I'm over the top. That's not how it went down. But they talk about the progression, the mile markers. What was the points where you guys had, "Okay, we achieved this, now let's start again?" Was there a pattern? Was there a progression that you could talk about?>> I think coming in, you can imagine when you arrive, and as Ed said, there was a huge... I always say to people there's a reason why I arrive at a job. It's not because everything's going really well and all you need to do is... I'm coming in so there's a problem, you need to fix it. We talked about the trust issue. The organization had lost trust in its ability to deliver. And so you come in and there's a whole bunch of stuff going on, very quickly realized that what we built, the cloud thing that we just built wasn't going to be the way forward model. So you're very early on looking at some enormous decision to say, "If I have to put that in containment..." Imagine coming into a new job, building credibility and you have to go, "I'm going to put that into containment." And so I think at that point there was a pilot running already that had been spun up before I arrived and I very quickly got into that. And I'm a super skeptical guy as Ed will tell you.>> So you came in, you had to do an assessment, you had to go in triage mode. Were you in that mindset?>> I've got people issues, I've got tech issues, I've got demand issues. I'm trying to sort through it all. And I remember the room I was in, I could tell you where it was, but I walked into the room and we had a Databricks briefing for me. I'm really hard guided to impress. And I went, "There is something here." And then you go, "Okay, I might have a light on the hill here. Of the many things I'm going to try and solve, I might have something here that I can work with."
And then you keep digging and as you get closer and closer to it, you think, "Actually, there's something real here." If you think about all the things we wanted to do, did it tick absolutely every box? No, but it ticked a lot of boxes and you meet Ed and Adam, I think was the account manager, and you start to meet the senior guys and you start to say, "Actually there's something here that's-">> They can pull it up. They get the capability.>> , right?>> When was the burn the boats moment or your turn, but when was the we're going all in moment? When you realized, "Okay, this is the path," was there an epiphany? Did you fall out of your chair? You're in the shower? You're on a hike?>> Once I'm convinced, I tend to make decisions relatively quickly. There was a burn the boats moment for getting off Teradata. There was a point where people said, "I don't think we can do this and Databricks is..." And I'm like, "We're done. Tell them we're not renewing the contract. There will be..." And the forcing issue there was great because it really forced us to push hard. And once we knew there was no contractual cover for the expiry of Teradata->> You thought you had people listed. Now the mob comes out and you weren't the most popular guy at that point, probably.>> No. There's an interesting story. We got a new account, Teradata account manager, I'm not sure if I'm allowed to say this, I will anyway, but he wrote an email to the CEO and said, "Your guys are crazy. They don't know what they're doing. They're putting your bank at all sorts of risks. It's out of control.">> That's a FUD bomb, basically, fear, uncertainty and doubt, scared, cognitive dissonance.>> Absolutely. That was the highlight of my whole process was getting the email. Then I got the email from the boss, we call him the boss, the CEO that said, "This is for you to deal with. Thank you.">> Ed, stories drive movements. This is a great transformation story. From your perspective, Databricks, this is what you do, how's everything else going on? Obviously this is a big customer here, but this seems to be the story. You got to take chances. You got to understand risk, you got to see the path, you got to know your data, you got to know your situation.>> And you got to have great clarity of purpose. So business is great across Asia Pacific and Japan. We love the work that we're doing with the National Australia Bank, but it turns out markets in Japan and Korea and India and Southeast Asia are all leaning into this in a big way. So we feel very, very lucky to work in these markets.>> And a lot of the country regulations play nicely into your hand. You're saying about the sovereignty issues around Unity Catalog is almost a secret superpower to manage the complexity of what the hell is coming next.>> So in Asia, for sure, we are meeting our customers where they are. They want us to be in their local countries. And so we have Databricks, as you know on AWS, Azure, GCP in each one of those countries and we make sure that we serve them. But, for sure, financial services often because of the regulatory requirements is our fastest growing vertical.>> How is the growth in APJ for you guys? What's the business look like? Give a quick minute, talk about some of the numbers, share some stats if you can.>> So the business is great. It's a pretty simple business. We don't have a demand constraint. And so it's all about fixing the supply side of the business. The reality is, over the next few years, every single organization in Asia is going to move to a single unified cloud-based data platform where AI is a first-class citizen. That's almost become an inevitability around modern data architectures. And so my job is to say, "Which markets are we going to prioritize? And then how do we make sure we've got all of the supply side capabilities in place to take on the seriousness of customers like the National Australia Bank and deliver them great outcomes over multiple years?" And so we've been hiring huge numbers of people to do that. We've built out organizations in Korea, in Japan, in India, in Australia, New Zealand, in Southeast Asia, and we just see a lot of growth ahead.>> It's nice when you get a nice business model that's simple to understand, to replicate and repeat.>> John, at the end of the day, I only do two things. I acquire customers and I grow customers. I have a very simple job. I just happened to do it in a fantastic part of the world.>> Christian, you got a great partner in Databricks. What's next for you guys? What's on your agenda? What's the focus?>> Well, obviously gen AI, agentic AI is a huge thing for us as it is for everybody. And, what, it's been around for a couple of years now and you think every time you step back you think, "Okay, I've got a handle on this," something else happens and it kicks up another gear. And so staying on top of that is probably a big issue for us at the moment. We are going to continue hydrating the platform. We want to get all of our data onto it and finish the job there. That's a big challenge for us. And really, I've said to my team now, because I run all the analytics, I do the producer and the consumer side, we actually need to really start pivoting towards on the consumer side, getting the business now thinking about how do you run the business differently? How do you change the game in terms of using data? Personalization, we've got a big initiative underway there that's really kicking goals. So all of that stuff now is we're shifting from producer, get all the data in that's still going, but now how do you consume the data and really start to power up the bank?>> So you're going to do a lot of unlocking, get this out there, get the creativity.>> But also, education. I use the example of Moneyball where baseball managers forever thought they knew how to play, how baseball worked and what was valuable and what was not. And then somebody came along with data and said, "Okay, well, actually that what you think works doesn't actually work the way you think it does.">> And someone on the team says, "Was this related to cricket?">> Yeah.>> They still have a salary cap too. Gentlemen, thank you for coming on. I really appreciate it.>> It's been absolute pleasure.>> Again, congratulations, Christian. Thanks for the story.>> No worries.>> Again, stories drive moving. The story here is transformation. Data like electricity in the banking area. It's about utility, but also growth, business growth, and user value. I'm John for theCUBE. We'll be back with more coverage from our full day of coverage here after the short break.
