This discussion explores applying knowledge graphs and artificial intelligence in financial services. John Furrier of theCUBE Research and Dave Vellante of theCUBE Research host a session recorded at Neo4j GraphSummit 2026 featuring Naveen Mahankali of Raymond James, vice president, technology, and Jeff Griffith of Raymond James, vice president, artificial intelligence engineering and data science.
Mahankali explains that agents and knowledge graphs enable advisors to query dynamic cross-domain information while preserving entitlements. They emphasize the need to constrain models to trusted data and to measure accuracy as part of responsible deployment.
Griffith describes how AI accelerates development velocity and shifts roles toward engineering and agent management. They highlight measurable productivity gains and vendor consolidation as drivers of adoption in financial services.
The conversation covers architecture decisions, integrating Neo4j, knowledge and context layers, graph-driven approaches that accelerate advisor workflows and enable cross-domain queries, and factors to consider such as data governance and data security.
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
theCUBE + NYSE Wired: Data + AI: Turning Data Into Knowledge for Autonomous Systems. If you don’t think you received an email check your
spam folder.
Sign in to theCUBE + NYSE Wired: Data + AI: Turning Data Into Knowledge for Autonomous Systems.
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 the link to automatically sign into the site.
Register for theCUBE + NYSE Wired: Data + AI: Turning Data Into Knowledge for Autonomous Systems
Please fill out the information below. You will receive 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 theCUBE + NYSE Wired: Data + AI: Turning Data Into Knowledge for Autonomous Systems.
Thanks for confirming your account. Now you can access theCUBE + NYSE Wired: Data + AI: Turning Data Into Knowledge for Autonomous Systems with this email address.
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
theCUBE + NYSE Wired: Data + AI: Turning Data Into Knowledge for Autonomous Systems. If you don’t think you received an email check your
spam folder.
Sign in to theCUBE + NYSE Wired: Data + AI: Turning Data Into Knowledge for Autonomous Systems.
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 the link to automatically sign into the site.
Sign in to gain access to theCUBE + NYSE Wired: Data + AI: Turning Data Into Knowledge for Autonomous Systems
Please sign in with LinkedIn to continue to theCUBE + NYSE Wired: Data + AI: Turning Data Into Knowledge for Autonomous Systems. 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
Jeff Griffith & Naveen Mahankali, Raymond James | Neo4j GraphTalk 2026
This discussion explores applying knowledge graphs and artificial intelligence in financial services. John Furrier of theCUBE Research and Dave Vellante of theCUBE Research host a session recorded at Neo4j GraphSummit 2026 featuring Naveen Mahankali of Raymond James, vice president, technology, and Jeff Griffith of Raymond James, vice president, artificial intelligence engineering and data science.
Mahankali explains that agents and knowledge graphs enable advisors to query dynamic cross-domain information while preserving entitlements. They emphasize the need to constrain models to trusted data and to measure accuracy as part of responsible deployment.
Griffith describes how AI accelerates development velocity and shifts roles toward engineering and agent management. They highlight measurable productivity gains and vendor consolidation as drivers of adoption in financial services.
The conversation covers architecture decisions, integrating Neo4j, knowledge and context layers, graph-driven approaches that accelerate advisor workflows and enable cross-domain queries, and factors to consider such as data governance and data security.
Jeff Griffith & Naveen Mahankali, Raymond James | Neo4j GraphTalk 2026
search
John Furrier
>> Palo Alto Studio Connection, Silicon Valley and Wall Street. I'm John Furrier, co-host of theCUBE, here with Dave Vellante, my co-host. Well, I'm John Furrier, host of theCUBE. We're back here at the Neo4j GraphSummit New York City. Two months ago, we were in San Francisco for the GraphSummit there, talking to the leaders, the practitioners, the partners, bringing in graph and knowledge graphs and new ways to bring AI into the applications and the value creation. We got a great customer here, practitioner Raymond James, very well known for their technology. We got two tech leaders here. We got Naveen, Vice President of Technology. Naveen, great to see you. Thanks for coming on.
Naveen Mahankali
>> Great to see
John Furrier
>> you.Jeff is the VP of AI Engineering and Data Science. Between the both of you, we're going to really unpack the AI opportunity. Thanks for coming on.
Jeff Griffith
>> Great to be
John Furrier
>> here.All right. So first question is, obviously, all the hubbub about AI, put that aside, all the narrative, a lot of noise in the market. There's real value coming out of the way you organize the data right now. We're starting to see graphs and ontologies going mainstream. You're starting to see the combination of other platforms, whether they're data lakes, come into play. You're starting to see an architecture of how to implement AI across organizations. So I guess my first question is, what are you guys doing?
Jeff Griffith
>> Well, we're doing a lot. There's a lot of different use cases. But I think there's a lot of commercial tools that are great that people can use for general uses. We're really focused on specific context to our business and bringing those capabilities that are really powerful to bear for our advisors and our associates in unique ways that apply to Raymond James.
John Furrier
>> Have you seen the user behavior shift from the analytics behavior, which is, hey, here's a bunch of dashboards, here's some reports, to much more real-time generative apps. And how do you guys think about it? Because the frontier models, they've crawled the internet. And so that's good if you want general intelligence, but if you want specialized intelligence, that's leveraging the data.
Jeff Griffith
>> Yeah, I think— go ahead.
Naveen Mahankali
>> Yeah, so one of the things we're trying to do is we're letting AI query the data dynamically so that FAs, our financial advisors, get what they want without clicking on screens and without looking at a fixed dashboard so they can do more analysis rather than just looking at the data that they already see in other applications. And this lets them connect data between different domains. Like they can do product analysis also versus how accounts are rebalanced and things like that by asking a question rather than going to multiple applications and getting the data and they have to manually tie that data together today.
John Furrier
>> So they're seeing value in the faster getting research product, which is basically their work product. Yeah. All right, so how do you guys do that? Are you guys leveraging knowledge graphs and knowledge layer context with Neo4j and other tools? How are you guys constructing the AI architecture?
Jeff Griffith
>> There's a lot of work we've done on unstructured data to really bring internal knowledge to enable people to use AI to get internal knowledge much faster, much easier. And the things that we're working on with data graphs are around the structured data, my clients, my positions, those things. And really there's a couple levers that you can pull. you want to use the best AI, you want to give it access to lots of information. It has to be the right information. So you have to be able to secure it. So your data is your data. And you have to be able to evaluate that the answers that you're giving are actually—
John Furrier
>> Yeah, it's huge. Or you can't
Jeff Griffith
>> have—If you can't measure it, then you can't deliver it.
John Furrier
>> Yeah.
Jeff Griffith
>> Yeah. So, those are really important things to us. Using something like a data graph lets us harness the power of AI in dynamic ways. People already have, we have a great platform. They already know how to go get the things that they're getting every day. But it's when you want to cross those domains or do some unique thing that this has a lot of value. So enabling AI to solve problems dynamically is really powerful. But like I said earlier, we have to be able to constrain it to the right entitlements and we want to limit the variables. So something like a data graph as opposed to just generating SQL, there's less variability there. The joins are already wired for you.
John Furrier
>> Yeah, I'm curious to get you guys' perspective because it's well known in the industry you guys are really strong on IT over the years. state of the art. As the AI wave came in, how did that hit you guys? okay, we have resilience, we got security posture, all the things you do. What was your frame when you framed AI? Because obviously you have customers, your customers and customers' customers, be the users and the customers. How did you guys look at this and how did you guys look at how it impacted the existing tech stack? Because again, you mentioned these queries, SQL queries, you have to do the intent and then do the query. Now you have the opportunity for agents on the other side to understand the prompt. So there's a lot of change on that kind of side of it. But given that you guys were already set up with the tech, How did you guys bring that AI in?
Jeff Griffith
>> I think it wasn't just a point in time thing. We've been doing all the flavors of AI through the years and for various purposes. And just like everyone else, we're adapting to each turn. And, those turns happen every 6 months or every month. Things are changing.
John Furrier
>> But that's a big change from last year to this year.
Jeff Griffith
>> The biggest thing that we've delivered is, uh, the ability to get really great knowledge from Knowledge, uh, without having to look it up on an intranet or to call someone for really unique and hard problems. There are other fit-for-purpose things that are rolling out for our advisors that are helping them plan meetings or do the work after meetings. And we have a really advisor-centric prioritization process. And so it comes down to what they're seeing value in and what they're pushing us to do. But overall, we want to give them the best tools and we want to make the
John Furrier
>> firm—a good impact so far.
Naveen Mahankali
>> We've seen a lot of acceleration in the last one year. Like, we have a big data science team, very good data science engineers, and Jeff himself built machine learning use cases for Raymond James. And in the last one year, we have taken advantage of how we build things using AI. So our team is equipped with building things faster now.
John Furrier
>> Right.Code assistants, vibe
Naveen Mahankali
>> coding.Yeah. So we are able to deliver in a much accelerated pace things that our financial advisors
John Furrier
>> need.It's interesting. I do so many interviews around AI and this area has been growing a lot, especially on the data platform. We've covered them all and it seems like AI has brought in these consequential benefits that weren't on the radar. obviously everyone wants to get some productivity, leverage the data, drives some high top-line impact or some value that can be quantified. One of the things that's come up, I'd love to get your reaction to it to see if you guys have the same experience, is that vendor consolidation actually was a big benefit. I'm like, well, that wasn't on my bingo card because AI can actually do a lot of things that other point solutions were doing with the intelligence. Are you seeing the relationship with your vendor suppliers change with AI? Does it change the role, the scope? Because when you use graphs or knowledge graphs, like a platform that has ontologies and knowledge and context layers, you go, whoa, I was using these tools, these bespoke tools.
Jeff Griffith
>> Well, maybe not a ton yet, but I think there is a huge shift in what you can make. The velocity of building software just, it does, kind of move the needle in terms of whether you want to go buy a product versus we can just make this. And I think what we see is AI driving our productivity statssort of off the charts right now.
John Furrier
>> Good. And you guys quantify that? Not yet, or? Well, it's advisory. So is that the main use case for that?
Jeff Griffith
>> Well, I would say it's quantified in the ways that technologists can quantify the number of changes to your software, the quality of—
John Furrier
>> So efficiency angle, pretty much it's that one. All right, here's a question for you guys because a lot of practitioners are jumping in, I would say maybe a little bit behind you guys on this front. The first question they ask is what's happening? And then it's what does it mean? And then what's next? Those are kind of the three questions. So I guess the question for you is, if you had to explain to a peer or your friend what's going on with this AI, how would you describe the what's happening right now in your world to them in their world? How would you explain that?
Naveen Mahankali
>> Yeah, so like I said, I think we've seen our engineers being more efficient and our testing methodologies are also very accelerated now. That we are able to deliver things in a month compared to 3 months in the past, right? If you take a typical use case. So that has helped us deliver significant impact for our financial advisors and now they are also seeing the benefit. And you want to add anything to that, Jeff?
Jeff Griffith
>> Yeah, well, it's a super exciting time because you have this new capability and It's getting smarter and smarter, but we're still absorbing all the ways that it can be used. And so you see it coming out in software development, you see it coming out in tools for the advisors. some people are afraid that the AI is going to do everything. But if you look at the world right now, how much does it really do? It does some nice things, but it certainly doesn't do everything. And you can live your day without it completely. So between here and doing everything, there's a lot of work that's gotta happen. And it's just—
John Furrier
>> everyone's impacted. AI scales intellect basically. to take the SQL query, that person's not out of a job, he's just out of a querying job. They will become data managers basically. And they know the data structures, they'll be more attuned to their performance. Versus grinding away and trying to figure out the intent for a static query.
Naveen Mahankali
>> Yeah, that is true. We've seen that actually, we, a lot of our engineers were coding in the past. Now they're truly doing the engineering work because we have delivered so much that we have to manage the agents and then we have to engineer the systems and scale. Yeah, that's what most of the work has shifted to.
John Furrier
>> Talk about the context graph, context layer, and then the knowledge graph. And how that impacts trust. Because, you know, on physical AI, you're seeing robotics, that's a pure safety issue on, you know, life or death or, you know, harm. And then on data accuracy, you don't want to get bad data and you don't want to have agents coming back with bad results that are pretty much the same thing. You have to have a solid, you know, no hallucination, zero tolerance for wrong answers.
Naveen Mahankali
>> Yeah, being a financial company, we need our answers to be accurate. We have golden sources, we have systems that we can rely on today, and we're able to use agents to build and compare what we are building against the golden sources. And we are able to measure the accuracy and improve the reliability of the agents that we are building. And with the knowledge graph also, even when we pull the data from different sources, it sits in the Neo4j And we run evals compared to the golden sources and we make sure they're 100% accurate before launching.
John Furrier
>> Yeah.
Jeff Griffith
>> The thing is, just to extend on to that, you can't deliver anything if you can't measure the accuracy in advance and ensure it. And then data security in terms of we're providing only the context that this person has access to, whether it's client data or it's just an article that is outside of their access. So putting those patterns in place is to ensure that is really important. I think another thing that we've done is internally we've tried to just stack the bricks. So we build great knowledge capabilities and we don't build them to solve one problem. We reuse that to solve all that type of problem. And there are other examples of that. And so that kind of brings us to a place where we're putting together pieces that are well known and trusted. And then when you improve a knowledge capability, you improve it for all users.
John Furrier
>> It's like getting smarter. Just adding more intelligence that compounds. It almost— I was talking with users from Neo4j earlier. It's like, if you asked me 3 years ago, 4 years ago, the data gravity problem would be solved or at least tackled at some level. I would've said, no way, it's going to be worse because all the forecasts were for more data. Obviously it's still coming in. Budgets aren't going up as fast as data. But when you introduce knowledge graphs and these trusted sources here and the way they're being architected is you can actually wrangle and tame the data gravity. What's your guys' take on that? Because you're living with a ton of data. Coming in every day?
Jeff Griffith
>> Yeah.
John Furrier
>> Being tamed or is it just kind of abstracted away?
Jeff Griffith
>> We're able to process way more of the signal than we used to be able to process. And we're able to turn a lot of things that are happening in our systems and with our users into knowledge that is constantly making our systems smarter and better.
John Furrier
>> Yeah, this is good. I wanna just end on a plug for you guys because, and there's a question in there around skills. What do you guys have for an environment? Engineers are coming in, I would agree with you Naveen that you're seeing a lot more engineering 'cause a lot of that scaffolding and grind is being kind of automated, but the workflows are still out there. Engineers want to solve hard problems. So when you're hiring, what kind of environment are people coming into when they join Raymond James environment? What's it like there? What problems are you guys working on? What are some of the constraints? Could you share without giving any IP for the company? What is it like and what's the need right now for you guys from a talent standpoint?
Jeff Griffith
>> Well, I'll let Naveen talk some about some of the engineering, but I think our unique sweet spot is that we're a large firm. We can do big things, but we don't have divisions headquartered all over, different places. So we can agree on what we're going to do and we can focus on some big things. And that's, that's what we've been doing. We have an AI assistant that I mentioned earlier that has— it's gotten huge adoption and it's just going to get smarter. And that's part of the core evolution of what we're doing in terms of engineering. We're changing how people— how everyone works with really a desktop that gives people the best AI tools to move from doing all the menial tasks to managing agents that are doing those tasks. And in terms of skills and what we're looking for, it is shifting just from people who can do tasks to people who have a drive to solve a meaningful problem. And with that comes the opportunity to solve that problem and to see people happy about that being solved. So there's a lot more opportunity, but you have to have the fortitude that you're going to solve something bigger than just completing tasks.
Naveen Mahankali
>> Yeah, just to add to that, I think Jeff covered most of it. Even though we are a big company, we are very nimble. So we are able to provide the best tools for our engineers, like anything that you can think of today, like Claude Code or anything we have available for our engineers, which some of the big companies I know are not making those available because of whatever concerns, right? So that actually changes our engineers' mentality and they're able to solve problems that would take years sometimes and I heard this from the engineers directly that they were able to solve things in a couple of weeks. And anybody who comes to us, they get that advantage to work and be nimble.
John Furrier
>> I use the word technology and engineering for a reason because IT used to be, IT. Now it's engineering, technology, IT, but it's business transformation going on. It's not an IT project. To your point about your adoption, probably saw that impact in the business. So we've seen IT everywhere now. So IT is now kind of standard in every part of the business, but the outcome seems to be business transformation, not an IT transformation. So it's really interesting to see, interesting that you guys share. And then the question is, okay, what kind of people do you hire? Because remember, the investments you guys are making are probably significant. So you got to do the data engineering, you got to get everything right. It's a very— it's a very fun time, but it's a generational architecture being built.
Jeff Griffith
>> It is very fun. You know what I see? I see people working on their own time a lot more, at their own choice. And I think whenever you would have a project that's going to— I'm thinking of something that we were trying to do and it was planned out without AI for like 4 months and we worked together and they were able to get that phase of it done in maybe 4 days. That work taking 4 months gets kind of boring, but when you compress it into a series of days, it's exciting to be able to make that much progress. And I think people love it.
John Furrier
>> They're having an acceleration earlier. That's the word, right? Accelerated.
Naveen Mahankali
>> One of the things I noticed is if I look at our team, we have data scientists, data engineers, infrastructure team, they're all able to cross paths now because with AI they can get visibility into what's happening in the other world. So that actually improved the productivity across teams
John Furrier
>> also.it's cliché to say teamwork, right? But there you're seeing evidence now, even at the C-suite level. The CHRO is working with the CFO. I've never seen that before. Usually it's a CIO and the CISO. And in engineering, you have cross-discipline activity because they have to work together because the data is horizontal, but the domain-specific work's good. So you have that old silo model, but they can still be silo-like, but they got to kind of play together. There's been a new kind of dynamic.
Jeff Griffith
>> Definitely within technology, the collaboration is increasing as the pace is.
John Furrier
>> from months to days. let's go get some beer after, get coffee, whatever people do. It's a fun time, guys. Thanks for sharing so much about you guys and appreciate it.
Jeff Griffith
>> Thanks for having us.
Naveen Mahankali
>> Yeah, thank you.
John Furrier
>> I'm John Furrier here at Neo4j GraphSummit New York City. This revolution continues. The AI infrastructure is accelerating this next layer of data, data platforms. Intelligence and context and knowledge graphs all work with all the data. And that seems to be the value in the AI equation. More coverage after this short break.
Jeff Griffith & Naveen Mahankali, Raymond James | Neo4j GraphTalk 2026
search
John Furrier
>> Palo Alto Studio Connection, Silicon Valley and Wall Street. I'm John Furrier, co-host of theCUBE, here with Dave Vellante, my co-host. Well, I'm John Furrier, host of theCUBE. We're back here at the Neo4j GraphSummit New York City. Two months ago, we were in San Francisco for the GraphSummit there, talking to the leaders, the practitioners, the partners, bringing in graph and knowledge graphs and new ways to bring AI into the applications and the value creation. We got a great customer here, practitioner Raymond James, very well known for their technology. We got two tech leaders here. We got Naveen, Vice President of Technology. Naveen, great to see you. Thanks for coming on.
Naveen Mahankali
>> Great to see
John Furrier
>> you.Jeff is the VP of AI Engineering and Data Science. Between the both of you, we're going to really unpack the AI opportunity. Thanks for coming on.
Jeff Griffith
>> Great to be
John Furrier
>> here.All right. So first question is, obviously, all the hubbub about AI, put that aside, all the narrative, a lot of noise in the market. There's real value coming out of the way you organize the data right now. We're starting to see graphs and ontologies going mainstream. You're starting to see the combination of other platforms, whether they're data lakes, come into play. You're starting to see an architecture of how to implement AI across organizations. So I guess my first question is, what are you guys doing?
Jeff Griffith
>> Well, we're doing a lot. There's a lot of different use cases. But I think there's a lot of commercial tools that are great that people can use for general uses. We're really focused on specific context to our business and bringing those capabilities that are really powerful to bear for our advisors and our associates in unique ways that apply to Raymond James.
John Furrier
>> Have you seen the user behavior shift from the analytics behavior, which is, hey, here's a bunch of dashboards, here's some reports, to much more real-time generative apps. And how do you guys think about it? Because the frontier models, they've crawled the internet. And so that's good if you want general intelligence, but if you want specialized intelligence, that's leveraging the data.
Jeff Griffith
>> Yeah, I think— go ahead.
Naveen Mahankali
>> Yeah, so one of the things we're trying to do is we're letting AI query the data dynamically so that FAs, our financial advisors, get what they want without clicking on screens and without looking at a fixed dashboard so they can do more analysis rather than just looking at the data that they already see in other applications. And this lets them connect data between different domains. Like they can do product analysis also versus how accounts are rebalanced and things like that by asking a question rather than going to multiple applications and getting the data and they have to manually tie that data together today.
John Furrier
>> So they're seeing value in the faster getting research product, which is basically their work product. Yeah. All right, so how do you guys do that? Are you guys leveraging knowledge graphs and knowledge layer context with Neo4j and other tools? How are you guys constructing the AI architecture?
Jeff Griffith
>> There's a lot of work we've done on unstructured data to really bring internal knowledge to enable people to use AI to get internal knowledge much faster, much easier. And the things that we're working on with data graphs are around the structured data, my clients, my positions, those things. And really there's a couple levers that you can pull. you want to use the best AI, you want to give it access to lots of information. It has to be the right information. So you have to be able to secure it. So your data is your data. And you have to be able to evaluate that the answers that you're giving are actually—
John Furrier
>> Yeah, it's huge. Or you can't
Jeff Griffith
>> have—If you can't measure it, then you can't deliver it.
John Furrier
>> Yeah.
Jeff Griffith
>> Yeah. So, those are really important things to us. Using something like a data graph lets us harness the power of AI in dynamic ways. People already have, we have a great platform. They already know how to go get the things that they're getting every day. But it's when you want to cross those domains or do some unique thing that this has a lot of value. So enabling AI to solve problems dynamically is really powerful. But like I said earlier, we have to be able to constrain it to the right entitlements and we want to limit the variables. So something like a data graph as opposed to just generating SQL, there's less variability there. The joins are already wired for you.
John Furrier
>> Yeah, I'm curious to get you guys' perspective because it's well known in the industry you guys are really strong on IT over the years. state of the art. As the AI wave came in, how did that hit you guys? okay, we have resilience, we got security posture, all the things you do. What was your frame when you framed AI? Because obviously you have customers, your customers and customers' customers, be the users and the customers. How did you guys look at this and how did you guys look at how it impacted the existing tech stack? Because again, you mentioned these queries, SQL queries, you have to do the intent and then do the query. Now you have the opportunity for agents on the other side to understand the prompt. So there's a lot of change on that kind of side of it. But given that you guys were already set up with the tech, How did you guys bring that AI in?
Jeff Griffith
>> I think it wasn't just a point in time thing. We've been doing all the flavors of AI through the years and for various purposes. And just like everyone else, we're adapting to each turn. And, those turns happen every 6 months or every month. Things are changing.
John Furrier
>> But that's a big change from last year to this year.
Jeff Griffith
>> The biggest thing that we've delivered is, uh, the ability to get really great knowledge from Knowledge, uh, without having to look it up on an intranet or to call someone for really unique and hard problems. There are other fit-for-purpose things that are rolling out for our advisors that are helping them plan meetings or do the work after meetings. And we have a really advisor-centric prioritization process. And so it comes down to what they're seeing value in and what they're pushing us to do. But overall, we want to give them the best tools and we want to make the
John Furrier
>> firm—a good impact so far.
Naveen Mahankali
>> We've seen a lot of acceleration in the last one year. Like, we have a big data science team, very good data science engineers, and Jeff himself built machine learning use cases for Raymond James. And in the last one year, we have taken advantage of how we build things using AI. So our team is equipped with building things faster now.
John Furrier
>> Right.Code assistants, vibe
Naveen Mahankali
>> coding.Yeah. So we are able to deliver in a much accelerated pace things that our financial advisors
John Furrier
>> need.It's interesting. I do so many interviews around AI and this area has been growing a lot, especially on the data platform. We've covered them all and it seems like AI has brought in these consequential benefits that weren't on the radar. obviously everyone wants to get some productivity, leverage the data, drives some high top-line impact or some value that can be quantified. One of the things that's come up, I'd love to get your reaction to it to see if you guys have the same experience, is that vendor consolidation actually was a big benefit. I'm like, well, that wasn't on my bingo card because AI can actually do a lot of things that other point solutions were doing with the intelligence. Are you seeing the relationship with your vendor suppliers change with AI? Does it change the role, the scope? Because when you use graphs or knowledge graphs, like a platform that has ontologies and knowledge and context layers, you go, whoa, I was using these tools, these bespoke tools.
Jeff Griffith
>> Well, maybe not a ton yet, but I think there is a huge shift in what you can make. The velocity of building software just, it does, kind of move the needle in terms of whether you want to go buy a product versus we can just make this. And I think what we see is AI driving our productivity statssort of off the charts right now.
John Furrier
>> Good. And you guys quantify that? Not yet, or? Well, it's advisory. So is that the main use case for that?
Jeff Griffith
>> Well, I would say it's quantified in the ways that technologists can quantify the number of changes to your software, the quality of—
John Furrier
>> So efficiency angle, pretty much it's that one. All right, here's a question for you guys because a lot of practitioners are jumping in, I would say maybe a little bit behind you guys on this front. The first question they ask is what's happening? And then it's what does it mean? And then what's next? Those are kind of the three questions. So I guess the question for you is, if you had to explain to a peer or your friend what's going on with this AI, how would you describe the what's happening right now in your world to them in their world? How would you explain that?
Naveen Mahankali
>> Yeah, so like I said, I think we've seen our engineers being more efficient and our testing methodologies are also very accelerated now. That we are able to deliver things in a month compared to 3 months in the past, right? If you take a typical use case. So that has helped us deliver significant impact for our financial advisors and now they are also seeing the benefit. And you want to add anything to that, Jeff?
Jeff Griffith
>> Yeah, well, it's a super exciting time because you have this new capability and It's getting smarter and smarter, but we're still absorbing all the ways that it can be used. And so you see it coming out in software development, you see it coming out in tools for the advisors. some people are afraid that the AI is going to do everything. But if you look at the world right now, how much does it really do? It does some nice things, but it certainly doesn't do everything. And you can live your day without it completely. So between here and doing everything, there's a lot of work that's gotta happen. And it's just—
John Furrier
>> everyone's impacted. AI scales intellect basically. to take the SQL query, that person's not out of a job, he's just out of a querying job. They will become data managers basically. And they know the data structures, they'll be more attuned to their performance. Versus grinding away and trying to figure out the intent for a static query.
Naveen Mahankali
>> Yeah, that is true. We've seen that actually, we, a lot of our engineers were coding in the past. Now they're truly doing the engineering work because we have delivered so much that we have to manage the agents and then we have to engineer the systems and scale. Yeah, that's what most of the work has shifted to.
John Furrier
>> Talk about the context graph, context layer, and then the knowledge graph. And how that impacts trust. Because, you know, on physical AI, you're seeing robotics, that's a pure safety issue on, you know, life or death or, you know, harm. And then on data accuracy, you don't want to get bad data and you don't want to have agents coming back with bad results that are pretty much the same thing. You have to have a solid, you know, no hallucination, zero tolerance for wrong answers.
Naveen Mahankali
>> Yeah, being a financial company, we need our answers to be accurate. We have golden sources, we have systems that we can rely on today, and we're able to use agents to build and compare what we are building against the golden sources. And we are able to measure the accuracy and improve the reliability of the agents that we are building. And with the knowledge graph also, even when we pull the data from different sources, it sits in the Neo4j And we run evals compared to the golden sources and we make sure they're 100% accurate before launching.
John Furrier
>> Yeah.
Jeff Griffith
>> The thing is, just to extend on to that, you can't deliver anything if you can't measure the accuracy in advance and ensure it. And then data security in terms of we're providing only the context that this person has access to, whether it's client data or it's just an article that is outside of their access. So putting those patterns in place is to ensure that is really important. I think another thing that we've done is internally we've tried to just stack the bricks. So we build great knowledge capabilities and we don't build them to solve one problem. We reuse that to solve all that type of problem. And there are other examples of that. And so that kind of brings us to a place where we're putting together pieces that are well known and trusted. And then when you improve a knowledge capability, you improve it for all users.
John Furrier
>> It's like getting smarter. Just adding more intelligence that compounds. It almost— I was talking with users from Neo4j earlier. It's like, if you asked me 3 years ago, 4 years ago, the data gravity problem would be solved or at least tackled at some level. I would've said, no way, it's going to be worse because all the forecasts were for more data. Obviously it's still coming in. Budgets aren't going up as fast as data. But when you introduce knowledge graphs and these trusted sources here and the way they're being architected is you can actually wrangle and tame the data gravity. What's your guys' take on that? Because you're living with a ton of data. Coming in every day?
Jeff Griffith
>> Yeah.
John Furrier
>> Being tamed or is it just kind of abstracted away?
Jeff Griffith
>> We're able to process way more of the signal than we used to be able to process. And we're able to turn a lot of things that are happening in our systems and with our users into knowledge that is constantly making our systems smarter and better.
John Furrier
>> Yeah, this is good. I wanna just end on a plug for you guys because, and there's a question in there around skills. What do you guys have for an environment? Engineers are coming in, I would agree with you Naveen that you're seeing a lot more engineering 'cause a lot of that scaffolding and grind is being kind of automated, but the workflows are still out there. Engineers want to solve hard problems. So when you're hiring, what kind of environment are people coming into when they join Raymond James environment? What's it like there? What problems are you guys working on? What are some of the constraints? Could you share without giving any IP for the company? What is it like and what's the need right now for you guys from a talent standpoint?
Jeff Griffith
>> Well, I'll let Naveen talk some about some of the engineering, but I think our unique sweet spot is that we're a large firm. We can do big things, but we don't have divisions headquartered all over, different places. So we can agree on what we're going to do and we can focus on some big things. And that's, that's what we've been doing. We have an AI assistant that I mentioned earlier that has— it's gotten huge adoption and it's just going to get smarter. And that's part of the core evolution of what we're doing in terms of engineering. We're changing how people— how everyone works with really a desktop that gives people the best AI tools to move from doing all the menial tasks to managing agents that are doing those tasks. And in terms of skills and what we're looking for, it is shifting just from people who can do tasks to people who have a drive to solve a meaningful problem. And with that comes the opportunity to solve that problem and to see people happy about that being solved. So there's a lot more opportunity, but you have to have the fortitude that you're going to solve something bigger than just completing tasks.
Naveen Mahankali
>> Yeah, just to add to that, I think Jeff covered most of it. Even though we are a big company, we are very nimble. So we are able to provide the best tools for our engineers, like anything that you can think of today, like Claude Code or anything we have available for our engineers, which some of the big companies I know are not making those available because of whatever concerns, right? So that actually changes our engineers' mentality and they're able to solve problems that would take years sometimes and I heard this from the engineers directly that they were able to solve things in a couple of weeks. And anybody who comes to us, they get that advantage to work and be nimble.
John Furrier
>> I use the word technology and engineering for a reason because IT used to be, IT. Now it's engineering, technology, IT, but it's business transformation going on. It's not an IT project. To your point about your adoption, probably saw that impact in the business. So we've seen IT everywhere now. So IT is now kind of standard in every part of the business, but the outcome seems to be business transformation, not an IT transformation. So it's really interesting to see, interesting that you guys share. And then the question is, okay, what kind of people do you hire? Because remember, the investments you guys are making are probably significant. So you got to do the data engineering, you got to get everything right. It's a very— it's a very fun time, but it's a generational architecture being built.
Jeff Griffith
>> It is very fun. You know what I see? I see people working on their own time a lot more, at their own choice. And I think whenever you would have a project that's going to— I'm thinking of something that we were trying to do and it was planned out without AI for like 4 months and we worked together and they were able to get that phase of it done in maybe 4 days. That work taking 4 months gets kind of boring, but when you compress it into a series of days, it's exciting to be able to make that much progress. And I think people love it.
John Furrier
>> They're having an acceleration earlier. That's the word, right? Accelerated.
Naveen Mahankali
>> One of the things I noticed is if I look at our team, we have data scientists, data engineers, infrastructure team, they're all able to cross paths now because with AI they can get visibility into what's happening in the other world. So that actually improved the productivity across teams
John Furrier
>> also.it's cliché to say teamwork, right? But there you're seeing evidence now, even at the C-suite level. The CHRO is working with the CFO. I've never seen that before. Usually it's a CIO and the CISO. And in engineering, you have cross-discipline activity because they have to work together because the data is horizontal, but the domain-specific work's good. So you have that old silo model, but they can still be silo-like, but they got to kind of play together. There's been a new kind of dynamic.
Jeff Griffith
>> Definitely within technology, the collaboration is increasing as the pace is.
John Furrier
>> from months to days. let's go get some beer after, get coffee, whatever people do. It's a fun time, guys. Thanks for sharing so much about you guys and appreciate it.
Jeff Griffith
>> Thanks for having us.
Naveen Mahankali
>> Yeah, thank you.
John Furrier
>> I'm John Furrier here at Neo4j GraphSummit New York City. This revolution continues. The AI infrastructure is accelerating this next layer of data, data platforms. Intelligence and context and knowledge graphs all work with all the data. And that seems to be the value in the AI equation. More coverage after this short break.
John Furrier
Naveen Mahankali
Jeff Griffith