Snowflake Data Cloud Summit 2024 | Aman Thind | State Street & Rinesh Patel | Snowflake

Clips
News
More from Snowflake Data Cloud Summit 2024

Aman Thind

Executive Vice President, Global Chief Architect

State Street

Rinesh Patel

Global Head of Financial Services

Snowflake

Data as the new development kit: Revolutionizing financial services to drive innovation

Once an ambitious declaration, the phrase “data is the new development kit” has now crystallized as the financial services sector’s reality.Given the current juxtaposition of macroeconomic volatilities and vast opportunities, how does a unified data approach, enabled by platforms such as Snowflake, help financial players separate the wheat from the chaff and deliver value where it matters?Experts from State Street and Snowflake discuss data transformation in the financial services sector

play_circle_outline Guests from State Street and Snowflake introduced

play_circle_outline The role of AI and large language models in application development

play_circle_outline The importance of responsible AI and trust in the financial services industry

play_circle_outline Partnership between State Street and Snowflake in driving innovation and serving clients' needs.

Info

Aman Thind | State Street & Rinesh Patel | Snowflake

Aman Thind

Executive Vice President, Global Chief Architect State Street

Rinesh Patel

Global Head of Financial Services Snowflake

search

Rebecca Knight

>> Good afternoon, everyone, and welcome back to theCUBE's live coverage of the Snowflake Data Cloud Summit, here at the Moscone Center. I'm your host, Rebecca Knight, sitting alongside my co-host and co-analyst, Dave Vellante. This is going to be a fun segment, talking about AI and its real impact on business.

>> Well, financial services always leads the trends. They're ahead of the game. They're technically very astute, so always excited to have the money folks on.

Rebecca Knight

>> Indeed. The money folks, I like it. Let me introduce our next guest. We have Aman Thind, he's the executive vice president, global chief architect at State Street. Welcome, Aman.

Aman Thind

>> Thank you for having me.

Rebecca Knight

>> And Rinesh Patel, global head of financial services at Snowflake. Thank you so much for coming back on theCUBE, Rinesh.

Rinesh Patel

>> Thank you for having me again.

Rebecca Knight

>> A CUBE veteran. So, I want to start with you, Aman. Why don't you tell us a little bit about your role at State Street and what you have responsibility for?

Aman Thind

>> I'm the chief architect for the firm. So, in that role, I run the technology strategy globally for State Street. And I'm also the chief technology officer for State Street Alpha. And Alpha is one of the industry's first front-to-back asset servicing platforms, which allows the investment managers to be able to get a single pane of glass and a single source of truth for all their data across the entire investment lifecycle and then be able to leverage that information to be able to make better-informed decisions.

>> And your scope is organization-wide or is it really more focused on the data platforms?

Aman Thind

>> State Street wide.

>> Okay. And of course, data is a feeder for everything, as we know.

Aman Thind

>> Exactly.

>> What's happening since the last year? A lot of changes in the marketplace, AI feeling a little bit more real, more products coming out. What are you seeing in the financial services industry in terms of adoption and excitement and caution?

Rinesh Patel

>> Yeah, all those things. A lot of change I think is probably the name of the game. If we think about the perspective of Aman's business focused on the buy side, the asset managers, the asset owners, we're seeing change from a macro perspective. When you think about higher interest rates, geopolitical challenges, volatility, I think that's certainly one area of challenge for the industry. The second is the investment process. Today's investors are trading more markets, trading more asset classes, being more systematic in their strategies, and the investors changing. Today's investors are more digitally-savvy, more financially-savvy. They're looking for deeper, timelier insights to make more decisions, hence the reason why data analytics is becoming very, very important to their decision making. And now, obviously with the rise of gen AI, we're seeing a better experience, a new experience being asked for by our customers and we're seeing obviously copilots, chatbots and so forth becoming very much mainstable within the industry.

>> So, Aman, this vision of a single version of the truth, we've been hearing it for decades. Are we finally at a point where we're going to be able to achieve that? And if so, how do you think about at least getting as close as possible?

Aman Thind

>> I think we're there, and the reason I say that is because there has been a seismic shift in terms of how applications are being written. Because historically, we would only write applications, and each application would come with its own data set. And what that meant was if you end up with 100 applications with slight varieties of those data sets, then you end up with a large distribution of your data and you'd never know where the truth truly lies. Now, with the advent of data platforms and all the innovation that is happening, for example at Snowflake, wherein it is compute that is coming to data and not data to compute, what that means is you can create a single platform which houses all your data sets, you bring all the analytics that you want to create, deploy them as containers on top of the data platform itself. And that allows you to process all the data in the single space and all your applications can be deployed as these single containers. So, this allows you to create multiple applications, finally, all accessing the same data set. And that, I think, is the true genesis of single source of truth. And I thank you guys for actually really leading the charge from that.

Rinesh Patel

>> Absolutely. Absolutely.

Rebecca Knight

>> So, I really want to get into what you were just talking about. With all of this macro political uncertainty, the number of countries that are choosing a new political leader this year. I mean, we just had elections in Mexico and India and of course the US-

Rinesh Patel

>> The UK.

Rebecca Knight

>> The UK, exactly. The US in five months from today. And these are sort of the global economic data that impacts institutions like the State Streets of the world. So, how is Snowflake innovating to make sure that it is keeping up with the changing data needs of these financial institutions?

Rinesh Patel

>> Yeah, I think there's an acceptance that there is a lot more data out there, unstructured, structured intel data, Excel data, a lot of third-party data. And from a Snowflake perspective, it's really about deep partnerships, partnering with those providers of those data, partnering with providers of platforms of data, and bringing that all together onto one network, i.e. our financial services data in the AI Cloud to enable customers to access strategically that data in a more seamless and frictionless way. And actually, we've been here in San Francisco for two days or so, and the amount of providers of data opining onto Snowflake, leveraging our network and making themselves available commercially. Those conversations have very much been topical for the last two days.

>> Now, so Aman, you mentioned something that resonated with me. We're now building applications with data, and we talk about the intelligent data apps and my business partner, John Furrier, years ago said, "Data is the new development kit," but it took a decade for us to get there. How do you see the way in which we develop applications changing? And I'm curious what role you see Snowflake playing in that.

Aman Thind

>> The gravity of data is now stronger than ever before. So, all your compute is moving towards data. And what that means is as your compute evolves more and more towards AI and large language models, all of that is moving towards data as well. So, all the innovation that is happening at Snowflake with everything that is being done around Document AI, Cortex, Arctic, what all of that means is it allows us to start leveraging AI, large language models, generative AI and our own traditional machine learning models. We can start deploying them on Snowflake itself. And this is now the new model of creating applications. Eventually, everything is just analytics being deployed on top of data sets, which you can visualize through low-code, no-code platforms.

>> So, when you talk about data gravity, help me square this circle because data is distributed by its very nature. And so, when you hear things like open table formats and Snowflake leaning in into that, this is data that's unfortunately outside of Snowflake, it's not all inside of Snowflake. So, how do you think about that as an architect? Do you say, "All right, for certain data where we absolutely need to have a single version of the truth, we'll bring that into Snowflake and develop apps on top of that. And for the rest of it we'll manage that"? How do you think about that?

Aman Thind

>> The reality is you're absolutely right. However, there are now technologies that are available that allows us to stitch all the data together. So, irrespective of whether you've got structured data in Snowflake tables or you've got unstructured data in PDFs or you've got data stuck in one box and drives and SharePoint. Now, using the latest technologies, you can take all of those data sets, you can create interrelationship between those data sets and you can create visualizations on top of that, you can create analytics on top of that without ever having to truly merge those data sets together. So, what that means is you don't have to have all your data in one place to be able to make the most of it. You can let the data sit where it is. But as long as your technology allows you to interrogate that data set, generate insights from that data set, and allows you to create the overall interrelationship between them, that is what truly unlocks the potential for the next leap in innovation and financial services.

>> And when you say visualize, I can think of three things. I can think of a traditional BI, I could think of a knowledge graph or I could think of some of more modern approach to democratized data. How do you think about it?

Aman Thind

>> It is all of them and none of them at the same time. And by that what I mean is all you need is when you ask a question, you want an answer. You don't want to chart, you don't want to get an application for which you need to go to tab number seven, and then run a filter and then get that information. And that is what large language models and chatbots and all the innovation that is happening in generative AI is providing us. So, what that means is now finally you are able to talk to your data. And as long as the data sets that run behind that power all of your artificial intelligence, it is agnostic to where the data comes from, how the data is stored, but it is able to understand the data sets, the meaning, their relationship, you can just talk to that. And then, you can ask your models to show it to you as a pie chart, as a bar chart, as a line graph or just give you a number, or create advanced analytics on top of that. And all of them are possible using all of those combinations you, or if you just want an answer, just ask and you shall receive.

Rebecca Knight

>> So, we know that the pace of change is dizzying and as Sridhar has said, the era of enterprise AI has arrived. Rinesh, from the horse's mouth, why is it so critical for financial institutions like State Street to have a robust strategy when it comes to enterprise data?

Rinesh Patel

>> I think it's essential. Like I said in the beginning, their customers are changing internally and externally, they're asking more and more complex questions around data analytics. So, to be able to provide them those answers at the right moment in time is really important. So, from a Snowflake perspective, it's really about bringing all that data together, internal, external, structured, unstructured together, enabling an organization to join that day together, as Aman mentioned. And then, providing an element of self-service, self-service back to that organization so they can enable those stakeholders to make those decisions at that timely point in time. So, from a Snowflake perspective, the platform overlaid with the ecosystem. And now, if you think about the selection of large language models we now have available to the platform, and as Aman just talked about, being able to create applications that talk to that data in natural language search and give you an answer back in natural language really brings true value to a customer in terms of answering those questions.

>> So, what are the challenges that you see in doing that? I mean a lot of customers will tell us they're being extra careful. Unlike some of the previous days of big data and there were a lot of shadow big data projects, and then two years later, governance came in. Not the case these days. How do you think about the challenges? What are those challenges in terms of deploying gen AI and how are you managing that?

Aman Thind

>> I think gone are the days when we used to say, "Move fast and break things," because now when you break things it is your customer's confidence and trust that you're breaking, and that is absolutely irreparable. So, the most important thing is to do anything that you are doing responsibly. And that is why at State Street we have a huge focus on a responsible AI framework and making sure that we are using the right technologies in the right way, while honoring all the data privacy rules, while honoring all the copyright laws. And just making sure that no matter what we're doing, it is absolutely rock solid in terms of its foundation around legal, around compliance, around risk. Because the biggest advantage of doing business with a brand name like State Street is that if you are doing that business with State Street, it comes with a notion of trust. And just doing that and helping make sure that we operate on that with velvet gloves is absolutely important. And that's where responsible AI and responsible AI framework comes in.

>> Move fast, but don't break things. But you still have to move fast, so how do you move fast without breaking things?

Aman Thind

>> So, the most important thing here is to be able to lean onto your partners because if we can't move fast enough, we can move with partners who want to move fast enough. So, in this case, that is where the beauty of the partnership between State Street, State Street Alpha and Snowflake truly lies wherein we are working with Snowflake on all the innovation in AI that is happening within that space, and they're helping us move faster and they're helping us propel forward while we bring in all the knowledge and all the experience around financial services, financial services data management and the needs of our clients and our 100-year experience with regards to managing these services, Snowflake brings in all the technologies and when you bring them together, yes, you can move fast without breaking anything.

Rinesh Patel

>> And then, take it to the customer. And take that actual partnership to the customers. Aman, we were discussing, we've met customers in London, we've met customers over in San Francisco, in Singapore and a few places. And taking that innovation and that partnership to our customers is a core part of the value as well.

Aman Thind

>> In fact, I had to shave so that people can actually tell the difference between Rinesh and I because we are meeting the same customers at the same, time talking about the same things.

Rebecca Knight

>> But also, I'm sure that it also builds trust, so that it's the partnership and customers can feel better. Okay, there are checks and balances happening here and the responsible AI, as you were talking about. I'm interested to hear the impact on teams. And in terms of financial services, there's this idea of we build it in-house. This is what we do in-house because then there's a certain pride of ownership too. Has there been a change in the mindset and the shift in how people think about their roles in your organization at State Street, the technologists that are working for you?

Aman Thind

>> The most important thing at State Street is its clients, and we work our way backwards from there. How can we provide value to our clients and the clients of our clients in the fastest, most efficient, most reliable, most scalable manner? And that is the guiding principle that is the North Star that we need to move towards and we will do anything and everything that we can in order to meet that need. And whether that means building something in-house or whether that means partnering with a client, partnering with a FinTech, no matter what it takes, but as long as you have, as an organization, one North Star, all these decisions become very easy to make.

>> Would you build your own LLM?

Aman Thind

>> No.

>> But there are people in your business that are contemplating that, are actually actively moving that direction, which-

Aman Thind

>> And I'm pretty sure they are. However, what we are focused on is not building that LLM, but actually using the best-in-class LLM in the market in order to service our clients. So, anyone who is trying to build an LLM is then subsequently bought into only using that LLM for servicing their clients. And then, what that means is it can be great at some things and it can be terrible at others. And if both those cases you are trying to use it for your clients, I think you're doing them a huge disservice. While at State Street, what we are doing is we are looking at the entire breadth of large-language models, everything that is available to us, and we are testing each and every one of them against all the things that we want to do. Do we want to summarize financial reports? Do we want to create text to SQL? Do we want to create a new code? Do we want to do code generation and commenting and testing? So, no matter what is needed, we test these language models against them, we score them, and then we fine-tune them and just make them better at doing what we need to. So, there will not be a single large-language model that rules them all. The industry is moving towards a mixture of experts and that is what we need.

>> I would agree, by the way, but that layer that you just described is that IP. You become experts at understanding which model and how to apply these models for your customers and your business. That's a value add.

Aman Thind

>> That's an immense value add because everyone is wondering where they should put their money on, but what they need to do is just put their money on State Street, and we will place the right bets on their behalf.

>> Plus, you got great companies like Snowflake and Meta building their own LLMs and spending all this CapEx. Beautiful. Thank you.

Rinesh Patel

>> Exactly right. Look, I think from our perspective, just echoing what Aman has said, very consistent. Rather than betting on a single horse in a rapidly-evolving landscape, we're extending our best-of-breed approach from clouds, to data providers, to now large language models. So, whether it's open source, external, or proprietary large language models, that's what we bring to customers because they're going to be wanting to solve or domain-specific problems or broad problems, and that's the value of what the partnership will bring.

Rebecca Knight

>> So, what's next for the partnership?

Rinesh Patel

>> Technically, a lot of conversations around how Arctic Cortex search can be embedded in the enterprise data strategy at State Street, but also, how we could take that innovation by way of State Street Alpha and some of those experiences to our customers.

Rebecca Knight

>> Excellent.

>> You guys should meet at the pub in London. I understand you both live in London, you don't see each other there. You come to here-

Rebecca Knight

>> You got to go to San Francisco to see this guy, yeah....

>> Las Vegas.

Rinesh Patel

>> I think we've definitely got to do that.

Aman Thind

>> Yeah, I think-

Rebecca Knight

>> All right. It's a date. I'm on Rinesh.

>> Or J Sheekeys.

Rebecca Knight

>> Thank you both so much for coming on theCUBE. A really great conversation.

Rinesh Patel

>> Thank you so much.

>> Thanks, you guys.

Rebecca Knight

>> I'm Rebecca Knight for Dave Vellante. Stay tuned for more of theCUBE's live coverage of the Snowflake Data Cloud Summit. You're watching theCUBE, the leader in enterprise tech news and analysis.

Aman Thind | State Street & Rinesh Patel | Snowflake

search

Rebecca Knight

>> Good afternoon, everyone, and welcome back to theCUBE's live coverage of the Snowflake Data Cloud Summit, here at the Moscone Center. I'm your host, Rebecca Knight, sitting alongside my co-host and co-analyst, Dave Vellante. This is going to be a fun segment, talking about AI and its real impact on business.

>> Well, financial services always leads the trends. They're ahead of the game. They're technically very astute, so always excited to have the money folks on.

Rebecca Knight

>> Indeed. The money folks, I like it. Let me introduce our next guest. We have Aman Thind, he's the executive vice president, global chief architect at State Street. Welcome, Aman.

Aman Thind

>> Thank you for having me.

Rebecca Knight

>> And Rinesh Patel, global head of financial services at Snowflake. Thank you so much for coming back on theCUBE, Rinesh.

Rinesh Patel

>> Thank you for having me again.

Rebecca Knight

>> A CUBE veteran. So, I want to start with you, Aman. Why don't you tell us a little bit about your role at State Street and what you have responsibility for?

Aman Thind

>> I'm the chief architect for the firm. So, in that role, I run the technology strategy globally for State Street. And I'm also the chief technology officer for State Street Alpha. And Alpha is one of the industry's first front-to-back asset servicing platforms, which allows the investment managers to be able to get a single pane of glass and a single source of truth for all their data across the entire investment lifecycle and then be able to leverage that information to be able to make better-informed decisions.

>> And your scope is organization-wide or is it really more focused on the data platforms?

Aman Thind

>> State Street wide.

>> Okay. And of course, data is a feeder for everything, as we know.

Aman Thind

>> Exactly.

>> What's happening since the last year? A lot of changes in the marketplace, AI feeling a little bit more real, more products coming out. What are you seeing in the financial services industry in terms of adoption and excitement and caution?

Rinesh Patel

>> Yeah, all those things. A lot of change I think is probably the name of the game. If we think about the perspective of Aman's business focused on the buy side, the asset managers, the asset owners, we're seeing change from a macro perspective. When you think about higher interest rates, geopolitical challenges, volatility, I think that's certainly one area of challenge for the industry. The second is the investment process. Today's investors are trading more markets, trading more asset classes, being more systematic in their strategies, and the investors changing. Today's investors are more digitally-savvy, more financially-savvy. They're looking for deeper, timelier insights to make more decisions, hence the reason why data analytics is becoming very, very important to their decision making. And now, obviously with the rise of gen AI, we're seeing a better experience, a new experience being asked for by our customers and we're seeing obviously copilots, chatbots and so forth becoming very much mainstable within the industry.

>> So, Aman, this vision of a single version of the truth, we've been hearing it for decades. Are we finally at a point where we're going to be able to achieve that? And if so, how do you think about at least getting as close as possible?

Aman Thind

>> I think we're there, and the reason I say that is because there has been a seismic shift in terms of how applications are being written. Because historically, we would only write applications, and each application would come with its own data set. And what that meant was if you end up with 100 applications with slight varieties of those data sets, then you end up with a large distribution of your data and you'd never know where the truth truly lies. Now, with the advent of data platforms and all the innovation that is happening, for example at Snowflake, wherein it is compute that is coming to data and not data to compute, what that means is you can create a single platform which houses all your data sets, you bring all the analytics that you want to create, deploy them as containers on top of the data platform itself. And that allows you to process all the data in the single space and all your applications can be deployed as these single containers. So, this allows you to create multiple applications, finally, all accessing the same data set. And that, I think, is the true genesis of single source of truth. And I thank you guys for actually really leading the charge from that.

Rinesh Patel

>> Absolutely. Absolutely.

Rebecca Knight

>> So, I really want to get into what you were just talking about. With all of this macro political uncertainty, the number of countries that are choosing a new political leader this year. I mean, we just had elections in Mexico and India and of course the US-

Rinesh Patel

>> The UK.

Rebecca Knight

>> The UK, exactly. The US in five months from today. And these are sort of the global economic data that impacts institutions like the State Streets of the world. So, how is Snowflake innovating to make sure that it is keeping up with the changing data needs of these financial institutions?

Rinesh Patel

>> Yeah, I think there's an acceptance that there is a lot more data out there, unstructured, structured intel data, Excel data, a lot of third-party data. And from a Snowflake perspective, it's really about deep partnerships, partnering with those providers of those data, partnering with providers of platforms of data, and bringing that all together onto one network, i.e. our financial services data in the AI Cloud to enable customers to access strategically that data in a more seamless and frictionless way. And actually, we've been here in San Francisco for two days or so, and the amount of providers of data opining onto Snowflake, leveraging our network and making themselves available commercially. Those conversations have very much been topical for the last two days.

>> Now, so Aman, you mentioned something that resonated with me. We're now building applications with data, and we talk about the intelligent data apps and my business partner, John Furrier, years ago said, "Data is the new development kit," but it took a decade for us to get there. How do you see the way in which we develop applications changing? And I'm curious what role you see Snowflake playing in that.

Aman Thind

>> The gravity of data is now stronger than ever before. So, all your compute is moving towards data. And what that means is as your compute evolves more and more towards AI and large language models, all of that is moving towards data as well. So, all the innovation that is happening at Snowflake with everything that is being done around Document AI, Cortex, Arctic, what all of that means is it allows us to start leveraging AI, large language models, generative AI and our own traditional machine learning models. We can start deploying them on Snowflake itself. And this is now the new model of creating applications. Eventually, everything is just analytics being deployed on top of data sets, which you can visualize through low-code, no-code platforms.

>> So, when you talk about data gravity, help me square this circle because data is distributed by its very nature. And so, when you hear things like open table formats and Snowflake leaning in into that, this is data that's unfortunately outside of Snowflake, it's not all inside of Snowflake. So, how do you think about that as an architect? Do you say, "All right, for certain data where we absolutely need to have a single version of the truth, we'll bring that into Snowflake and develop apps on top of that. And for the rest of it we'll manage that"? How do you think about that?

Aman Thind

>> The reality is you're absolutely right. However, there are now technologies that are available that allows us to stitch all the data together. So, irrespective of whether you've got structured data in Snowflake tables or you've got unstructured data in PDFs or you've got data stuck in one box and drives and SharePoint. Now, using the latest technologies, you can take all of those data sets, you can create interrelationship between those data sets and you can create visualizations on top of that, you can create analytics on top of that without ever having to truly merge those data sets together. So, what that means is you don't have to have all your data in one place to be able to make the most of it. You can let the data sit where it is. But as long as your technology allows you to interrogate that data set, generate insights from that data set, and allows you to create the overall interrelationship between them, that is what truly unlocks the potential for the next leap in innovation and financial services.

>> And when you say visualize, I can think of three things. I can think of a traditional BI, I could think of a knowledge graph or I could think of some of more modern approach to democratized data. How do you think about it?

Aman Thind

>> It is all of them and none of them at the same time. And by that what I mean is all you need is when you ask a question, you want an answer. You don't want to chart, you don't want to get an application for which you need to go to tab number seven, and then run a filter and then get that information. And that is what large language models and chatbots and all the innovation that is happening in generative AI is providing us. So, what that means is now finally you are able to talk to your data. And as long as the data sets that run behind that power all of your artificial intelligence, it is agnostic to where the data comes from, how the data is stored, but it is able to understand the data sets, the meaning, their relationship, you can just talk to that. And then, you can ask your models to show it to you as a pie chart, as a bar chart, as a line graph or just give you a number, or create advanced analytics on top of that. And all of them are possible using all of those combinations you, or if you just want an answer, just ask and you shall receive.

Rebecca Knight

>> So, we know that the pace of change is dizzying and as Sridhar has said, the era of enterprise AI has arrived. Rinesh, from the horse's mouth, why is it so critical for financial institutions like State Street to have a robust strategy when it comes to enterprise data?

Rinesh Patel

>> I think it's essential. Like I said in the beginning, their customers are changing internally and externally, they're asking more and more complex questions around data analytics. So, to be able to provide them those answers at the right moment in time is really important. So, from a Snowflake perspective, it's really about bringing all that data together, internal, external, structured, unstructured together, enabling an organization to join that day together, as Aman mentioned. And then, providing an element of self-service, self-service back to that organization so they can enable those stakeholders to make those decisions at that timely point in time. So, from a Snowflake perspective, the platform overlaid with the ecosystem. And now, if you think about the selection of large language models we now have available to the platform, and as Aman just talked about, being able to create applications that talk to that data in natural language search and give you an answer back in natural language really brings true value to a customer in terms of answering those questions.

>> So, what are the challenges that you see in doing that? I mean a lot of customers will tell us they're being extra careful. Unlike some of the previous days of big data and there were a lot of shadow big data projects, and then two years later, governance came in. Not the case these days. How do you think about the challenges? What are those challenges in terms of deploying gen AI and how are you managing that?

Aman Thind

>> I think gone are the days when we used to say, "Move fast and break things," because now when you break things it is your customer's confidence and trust that you're breaking, and that is absolutely irreparable. So, the most important thing is to do anything that you are doing responsibly. And that is why at State Street we have a huge focus on a responsible AI framework and making sure that we are using the right technologies in the right way, while honoring all the data privacy rules, while honoring all the copyright laws. And just making sure that no matter what we're doing, it is absolutely rock solid in terms of its foundation around legal, around compliance, around risk. Because the biggest advantage of doing business with a brand name like State Street is that if you are doing that business with State Street, it comes with a notion of trust. And just doing that and helping make sure that we operate on that with velvet gloves is absolutely important. And that's where responsible AI and responsible AI framework comes in.

>> Move fast, but don't break things. But you still have to move fast, so how do you move fast without breaking things?

Aman Thind

>> So, the most important thing here is to be able to lean onto your partners because if we can't move fast enough, we can move with partners who want to move fast enough. So, in this case, that is where the beauty of the partnership between State Street, State Street Alpha and Snowflake truly lies wherein we are working with Snowflake on all the innovation in AI that is happening within that space, and they're helping us move faster and they're helping us propel forward while we bring in all the knowledge and all the experience around financial services, financial services data management and the needs of our clients and our 100-year experience with regards to managing these services, Snowflake brings in all the technologies and when you bring them together, yes, you can move fast without breaking anything.

Rinesh Patel

>> And then, take it to the customer. And take that actual partnership to the customers. Aman, we were discussing, we've met customers in London, we've met customers over in San Francisco, in Singapore and a few places. And taking that innovation and that partnership to our customers is a core part of the value as well.

Aman Thind

>> In fact, I had to shave so that people can actually tell the difference between Rinesh and I because we are meeting the same customers at the same, time talking about the same things.

Rebecca Knight

>> But also, I'm sure that it also builds trust, so that it's the partnership and customers can feel better. Okay, there are checks and balances happening here and the responsible AI, as you were talking about. I'm interested to hear the impact on teams. And in terms of financial services, there's this idea of we build it in-house. This is what we do in-house because then there's a certain pride of ownership too. Has there been a change in the mindset and the shift in how people think about their roles in your organization at State Street, the technologists that are working for you?

Aman Thind

>> The most important thing at State Street is its clients, and we work our way backwards from there. How can we provide value to our clients and the clients of our clients in the fastest, most efficient, most reliable, most scalable manner? And that is the guiding principle that is the North Star that we need to move towards and we will do anything and everything that we can in order to meet that need. And whether that means building something in-house or whether that means partnering with a client, partnering with a FinTech, no matter what it takes, but as long as you have, as an organization, one North Star, all these decisions become very easy to make.

>> Would you build your own LLM?

Aman Thind

>> No.

>> But there are people in your business that are contemplating that, are actually actively moving that direction, which-

Aman Thind

>> And I'm pretty sure they are. However, what we are focused on is not building that LLM, but actually using the best-in-class LLM in the market in order to service our clients. So, anyone who is trying to build an LLM is then subsequently bought into only using that LLM for servicing their clients. And then, what that means is it can be great at some things and it can be terrible at others. And if both those cases you are trying to use it for your clients, I think you're doing them a huge disservice. While at State Street, what we are doing is we are looking at the entire breadth of large-language models, everything that is available to us, and we are testing each and every one of them against all the things that we want to do. Do we want to summarize financial reports? Do we want to create text to SQL? Do we want to create a new code? Do we want to do code generation and commenting and testing? So, no matter what is needed, we test these language models against them, we score them, and then we fine-tune them and just make them better at doing what we need to. So, there will not be a single large-language model that rules them all. The industry is moving towards a mixture of experts and that is what we need.

>> I would agree, by the way, but that layer that you just described is that IP. You become experts at understanding which model and how to apply these models for your customers and your business. That's a value add.

Aman Thind

>> That's an immense value add because everyone is wondering where they should put their money on, but what they need to do is just put their money on State Street, and we will place the right bets on their behalf.

>> Plus, you got great companies like Snowflake and Meta building their own LLMs and spending all this CapEx. Beautiful. Thank you.

Rinesh Patel

>> Exactly right. Look, I think from our perspective, just echoing what Aman has said, very consistent. Rather than betting on a single horse in a rapidly-evolving landscape, we're extending our best-of-breed approach from clouds, to data providers, to now large language models. So, whether it's open source, external, or proprietary large language models, that's what we bring to customers because they're going to be wanting to solve or domain-specific problems or broad problems, and that's the value of what the partnership will bring.

Rebecca Knight

>> So, what's next for the partnership?

Rinesh Patel

>> Technically, a lot of conversations around how Arctic Cortex search can be embedded in the enterprise data strategy at State Street, but also, how we could take that innovation by way of State Street Alpha and some of those experiences to our customers.

Rebecca Knight

>> Excellent.

>> You guys should meet at the pub in London. I understand you both live in London, you don't see each other there. You come to here-

Rebecca Knight

>> You got to go to San Francisco to see this guy, yeah....

>> Las Vegas.

Rinesh Patel

>> I think we've definitely got to do that.

Aman Thind

>> Yeah, I think-

Rebecca Knight

>> All right. It's a date. I'm on Rinesh.

>> Or J Sheekeys.

Rebecca Knight

>> Thank you both so much for coming on theCUBE. A really great conversation.

Rinesh Patel

>> Thank you so much.

>> Thanks, you guys.

Rebecca Knight

>> I'm Rebecca Knight for Dave Vellante. Stay tuned for more of theCUBE's live coverage of the Snowflake Data Cloud Summit. You're watching theCUBE, the leader in enterprise tech news and analysis.