Data Analytics Chat

How Do We Thread Gen AI Into Our Processes?

• Ben Parker

Share episode

Send us a text

In this episode of Data Analytics Chat, host Ben Parker welcomes Tony Giordano, the Senior Partner and VP for Global Data Platform Services at IBM. They delve into Tony's extensive career, including pivotal moments and lessons learned from working globally in the data space. The conversation covers the transformative impact of Gen AI, the evolution of data architecture, and the strategic integration of AI into business processes. Tony shares valuable insights on navigating technological advancements, maintaining domain expertise, and preparing for a future driven by data innovations. Tune in for expert advice and real-world stories from the front-lines of data consulting.

00:00 Introduction to the Exciting World of Data Consulting
01:08 Welcome to Data Analytics Chat
01:45 Tony Giordano's Career Journey
05:31 The Impact of Working in Different Countries
08:48 Leading a Global Team at IBM
10:43 Cultural Lessons from Japan
22:56 Career Setbacks and Lessons Learned
24:45 The Future of Data and AI
28:58 The Calculator Debate and Modern Tools
30:31 The Importance of Domain Expertise
32:08 Celebrating Data Luminaries
33:29 The Evolution of Data Patterns
35:40 The Shiny Object Syndrome
39:48 The Role of Gen AI in Business
44:20 Cultural and Organizational Challenges
48:57 The Future of Data and Gen AI

Thank you for listening!

tony giordano:

The world we're in right now is in such flux that I can't think of a more exciting time to be in the data space and being in the consulting space because we as consultants are usually two to three chapters ahead of our constituent organizations, at least a chapter ahead. We're getting to do some of the cool stuff out in the business in this space called data. We are showing, proving at some of our client set the move from just even just traditional Gen AI, the quote unquote chat bot whole thing, but really creating agents to do things rather than just a Gen AI chat bot is gonna. Truly transform. Again, it's a second turn of the crank that clock. When I say change a clock, things that used to take a thousand hours, we've conservatively guessed can be 35 to 60%. We keep it at 35%'cause no one's gonna complain if you come in early. We don't wanna be too aggressive, but we know it's probably closer to 50.

ben parker:

Welcome to Data Analytics Chat, the podcast where we discuss the world of data AI and the careers shaping it. I'm your host, Ben Parker, bringing you real stories, expert insights, and practical advice to help you thrive in the data industry Today I'm excited to welcome Tony Giordano, the senior partner and vice president for Global. Data Platform services leader at IBM. In today's episode, we'll explore his career journey, pivotal moments that shaped his career path, and also discuss how do we thread Gen AI into our processes. Tony, welcome to the podcast.

tony giordano:

Ben, it's a pleasure to be here.

ben parker:

Brilliant. I'm looking forward to this. Obviously, Gen AI's obviously the popular topic, so it's gonna be an interesting conversation.

tony giordano:

I can't tell you how exciting it is to be in the data business right now on Twitter. Sitting here in 2025. Okay.

ben parker:

Let's dive in then. And do you wanna, I mean, do you wanna start with sharing your career journey?

tony giordano:

Sure. So my career journey has almost completely been about data. You know, I started, um, started really in university.'cause in university I, I paid for my university. So I needed a job and I got a job as a programmer. And, uh, it was just a lucky break. And I built commercial banking applications for about three and a half years while I was surfing through university. And when I graduated, I immediately went into the consulting business. I, uh, got an opportunity to work with a, a fellow British gentleman by the name of James Martin. He, uh, was really big at the time in a concept called information engineering. And you had a whole slew of books. Still some of my. Three favorite books, information Engine one, two, and three working for him. I got to work on a very small project. At the Internal Revenue Service where they were doing a project called Tax Systems Modernization, and it was taking things off the mainframe and putting'em on time in, in, uh, early client server. And I got to work with technology such as Oracle and Teradata and, a lot of the technologies we've been working in the data space for many, many years. And, uh, that led me to, uh, a job offer to Oracle and, uh, spent a bit of time working in London. In fact, I was in, well down in Churchy working for Oracle. Did that for about a year and a half, almost two years. And then, then I got a chance to go work for a global team called, pricewaterhouse Coopers or Pricewaterhouse at the time actually, and joined them. And I was a small band of about 50 practitioners that were really defining what data warehousing world was gonna be all about early aughts. So I joined that, uh, had a great time there. Uh, and never left to be honest. I stayed there, was focused on data and business intelligence and, and data integration ended up, uh. Staying through the IBM acquisition, ended up running North America data through about$450 million business. And then, um, at IBM, you, you have this thing, you, you won't get promoted unless you have got certain stripes and IBM stands for, I've been moved, so I took a, a job in, in Japan, so I ended up running the data practice in Japan for a couple years. Took a little walk about, I had a chance to work for a digital marketing company called Merkel. Went there then, uh, IBM called me back up and Ben, you seemed to be a pretty smart guy. Merkel was based in Maryland and I was living in Baltimore, and I was gonna leave Merkel and join a small company in Philadelphia. When, um, I got a call from my old friends at IBM, they said, Hey, we'd love you back. I said, ah, I've been there, done that. They said, yeah, but we need somebody to run our European practice. I said, okay, from where? And they said, well, from Paris. Faced with the choice of Philadelphia or Paris, I chose Paris. Came back to IBM, helped run the European practice, did that for a couple years and had a lot of fun. Really helped transform the data warehousing business into what's now really our open source, cloud-based data platform business that we have today. And, uh, we have a lot of success. And I was asked to take over the global role and that's when I took that over and we've grown it to well over a billion dollar business. That's a quick synopsis of my background, but as you can hear, my whole career has been around this, this, uh, really interesting space called data. Yep. And it's,

ben parker:

yeah, take going, taken off now, isn't it? It's just crazy. Uh, it's obviously amazing to be in as well. A lot, a lot happening. I've been, I guess obviously you've worked in some interesting countries. How, how do you think that shapes your career? Every

tony giordano:

single project experience, every single client, every single person I've had a chance to work with, whether in country or outta country, has formed my point of view on data and how hard or how easy it is to do. I will tell you a lot of my confidence is based on the facts, that once you see a pattern established in one place done by very different people,'cause the Japanese do not speak English very well, but if I can build. A third normal form database here, a dimensional model There, a de-normalized pattern on cloud object store over here. I happen to know it's, it's probably something I can do anywhere. Come to the conclusion over time there are certain patterns in data in it that are never changing, but constantly evolving. And that's really what I've observed over my career is there's a lot of things that we do end data to this day that. Go back to when I started in this business and some of the people I, I learned from, but it's much more evolved today than it was back then. Does that make

ben parker:

sense? Yeah, no, definitely. I think obviously things are, yeah, obviously constantly changing, aren't they? And I think, like I said you've been braver than me. I moved from the UK to Spain. But I think when you do move countries, you do, you meet new people, you do learn different perspectives. I think it does add to your personality and you learning, doesn't it?

tony giordano:

Well, it's interesting. I was literally making fun of myself yesterday, and I was presenting to a group of practitioners about. How to do solutioning in the age of, of ent, because a lot of folks don't realize how much solutioning is gonna change. And we don't need three months to build data models. We can do it in about 10 minutes. And I made some comment like, this ain't good. That somebody said, well, that's not very correct English. And I said, yeah, but I'm from the Midwest when I lived in Britain. You know, I lived there twice. They were always telling me I didn't speak English very well, so I, I just accepted that. And it's, it's a little fun, poking fun at oneself. But yes, the cultural changes from one country to the next have informed how I interact with people, and I found self-deprecating humor to be incredibly useful regardless of the audience and self-deprecating humor is good in, in Chicago, New York, Los Angeles, Hong Kong. Indonesia, and I've been very, Ben, I've been, I've had a very fortunate career. I've, I've literally worked across the planet. I've worked in Jakarta, Thailand, India. I was just over in India. I just over in Japan. Been in Europe three times this year. I was in Amsterdam twice. I was in Vienna once. So I've had a very. Fruitful career and very lucky.

ben parker:

Yeah, man, that's amazing.'cause I think, I think when you go to different like countries, it's new learnings. It's dealing with new people. I know. I know. Obviously everywhere you go now it's quite diverse. But when you've got the opportunity to work in different places, it does add to you. I mean, I've done sort of like remote work in done traveling in Asia and things like that. It is good to meet. Different cultures and things like that. I think it, it's interesting, um, and it, you, I said it does shape you, gives you confidence and to meet new people and understand people's ways of life as well.

tony giordano:

So, man, here, here's an interesting thing to share. I often forget about it while I run the global practice and there's about 7,000 of us, and that's, senior executives to, entry-level developers. I have a direct team. Of about 40 people. And of those 40 people, there's probably 10 of'em that are, are the people I go to to get a lot of stuff done. This is from offering management to deployment, to development, you know, to building assets we use for the way we do data today. This is one of the pivots IBM has really invested in is pivoting towards asset-based. Development, and we can get into that in just a minute, but just to let you know, the call I ran, I run every Friday morning and now I saw you're the best people I've ever had a chance to work with. And I mean it every single time from the heart, not from the head. My development leader is in, uh, North Carolina. Not that fancy Raleigh, but my deployment leader lives in Singapore. One of my, one of my, uh, integration specialists lives in Morocco. I've got other folks living in India. I've got folks in Europe. And, uh, Eslava. I got a group in Poland. So I run truly a global team on top of running a global practice. And oh, and my, my campaign people, God help me if I forget them'cause they're. Some of the most important people on my team all live in Barcelona. So every Friday morning we have a stew of different countries, but what we have is a bunch of common practitioners with great passion for what we do, who all speak the common language called data and serving our clients. And we have a hell of a lot of fun doing it.

ben parker:

And I think that's what matters. And it, if you've all got that. Shared goals. All it matters, doesn't it really? So then, I mean, for yourself then, Tony, you've obviously, you've, you're at IBM and then left and joined again. How was that?

tony giordano:

So what's interesting is, um, you, you brought up a interesting cultural thing. Where I think you've probably very quickly ascertained my personality. It's, it's pretty apparent, pretty bold, and I, I'm aware of who I am and my limitations and my strengths and my weaknesses. When I was asked to go to Japan, I was going, because I was asked to go to Japan for a couple D reasons. Number one, the data team in Japan had never made their numbers ever. Secondly, they had brought a, a lot of other American gaja over there to try it and, okay, this is Giordano's turn to try this thing. I had been to Japan on a couple other trips for other reasons, and I, I really sat back to reflect how am I gonna do this in a way that is successful and successful for me personally, successful for IBM, and most importantly, successful for the Japanese. So I made some ground rules. First ground rule was I am not going to try to be somebody. I'm not. I'm a Midwestern boy. I'm not ashamed nor proud. I just, it is who I am. I grew up in, in Michigan, my dad worked for General Motors. You know, I, I went to school in Ann Arbor. I'm, I'm a mid-westerner. Try to be something I'm not, would not make any sense. So I'm, I'm not gonna fit in. The Japanese are very insular, and I, I, I, it's funny, the last day I finally figured out difference between JI and Gaja and why we get called Gaja. If you're not of the people, then you're, you're outside the people. That doesn't mean you, you will never be accepted or you won't be treated with respect. It just means. You need to understand your place. So one of the things I observed, and this is really important because this gets into one of the cultural lessons I've learned. Nothing to do with data, but just how to do deal with people other places. Most Americans or Brits that would go to Japan to to, to live and work, they'd show up around 7 30, 8 in the morning and they would. Leave around five or six. Well, most Japanese people don't come that early. They usually show around 8 39. They leave at 1130 at night. And this is something that's been instilled in the Japanese culture back to the nineties when they gained their industrial eminence. They kind of to charge in the rural economy. And a lot of Japanese don't have to stay that late, but it's almost a shame if they leave early. It's like, why are you home so early so the Japanese don't leave till 1130? So I show up, it's the middle of August, actually, almost toward weather. We have right now, early August, and it was very, very hot over there. And this wasn't that long after f Kajima Japan was suffering power problems, but all Japanese companies did to, show solidarity with the Japanese government. After five o'clock, they would turn off the air conditioning in July when it's around, you know, 85, 90 degrees Fahrenheit. So my first day there, and back then the Japanese would still wear three piece suits and they walk around in three piece suits. If you're not at your desk, if you at your desk, you take your jacket off. So my first day there, I kept my jacket on even sitting there. And then five o'clock hits and you can start feeling the air conditioning go off six o'clock, seven o'clock, eight o'clock. The team I'm now responsible for keeps looking over me like they're waiting for me to leave. I wasn't gonna leave 10 11, I'm not leaving. My jacket's still on. It's probably 85 degrees in that office now. And, and some of the Japanese are not loosening their tie. I'm loosened my tie and I'm sitting there. I'm furiously working and I'm now back at North America time because keep in mind, this is my first day there and, um, about 1230 somebody goes. Tony's son are, are you gonna go home? I said, well, I'm gonna leave with the team. One o'clock, one 30, people start leaving quarter to two. The last person walks out and I walk out with them. Next night, we leave around 1230, and from then on they realized I was gonna be the last person out the door because I was gonna support the team. Now, for a team that never made their numbers. They would listen to me. I would say, here's what we need to do to sell this stuff. Here's how we need to do deliver it. Because most of these folks are still back in the mainframe day. Well, why aren't we doing this on IBM mainframe or IMS? I go, because the world is now, moved long, long beyond that. So I would spend sessions on Friday night teaching them the new patterns that we are doing. Where we've been successful with this.'cause the Japanese will not be the first in anything. They will replicate and improve on any pattern that has been successful elsewhere. Our first quarter, Ben, we made it by$2 million. The second quarter we made it by$12 million. But even on Fridays we left at 11 30, 12 o'clock. After two years of this, I was ready for something different and, um. We were making our numbers. The team was great. I got a phone call from a recruiter and they said, Hey, there's a small marketing company they're trying to grow called Merkel. It's all, it's a digital marketing agency. They would love to talk to you. We had a lot of chats. They reminded me a lot of Pricewaterhouse back in the days I first joined, and I said, you know, I might be fun to go do that again. And so I left IBM was hard to lead my team. They, they the farewell party was simply epic. They got videos from. People, I wouldn't imagine. And it, it was, uh, it was actually very sad. And, um, to this day, some of those people are some of my best friends now. But I left, I joined Merkel, moved to Baltimore, Maryland, and, uh, first 18 months were great. I, I learned so much that a lot of the MarTech people were further ahead than the IT people when it came to the use of data and the immediacy of data in the digital space. It was a, it was a humbling experience to realize how far behind IBM and a lot of consultancies, other consultancies were in the use of data, especially real-time data. So learned a lot. The owner was a great guy for various reasons. He had to sell the company. He chose to, to sell it, and had even just gone through an acquisition a few years before. I said I might wanna look other places, and that's when I took a, took a look, and I actually accepted a job from a little company in Philadelphia. The IBMO opportunity came up and like I said, Philly cheese steaks or, uh, beignets and, uh, coffee in, uh, Paris. So it was an easy, easy decision for me to make. Gave you a long-winded story, but you asked an important question,

ben parker:

Brian. I love that story and obviously it's yeah, great to see you lead by example. Basically,

tony giordano:

it's the only way you, the only way I've learned to really get a team to buy into what, what you're doing. If you're not willing to work as hard as they are and show the same passion they are, they'll only raise their passion. To what you're willing to contribute if I'm gonna give it up my all, and I've never seen a team not give it their all. So then looking back. Any key moments that shaped your career? I think every major decision I made in terms of moving jobs, moving from one company to the next were all major things that I, I could think of. A lot of engagements came up. It's like, what am I doing this, I would say I almost embraced the challenging projects more than the, than successful projects I've, I've had more than my fair share of really successful projects, things that are just. Textbook. Perfect. I've only had one project that's actually shut down and that's because a client lost funding and that it wasn't from the work that we were doing.'cause we were actually on time and on track. They just, they had their funding cut. I've had to work through a lot of challenging projects. In fact, for a while they would say, Hey, we got a fire over here. We need you to go put it out. And I became kind of an IBM firefighter to fix broken projects and that I've never seen a, a project that can't be fixed technically. I've seen messes that have been created that. Aren't fixable. Technology has never been the problem. Either the requirements were bad or unknown, or the people on the project just weren't aligned with the stakeholders and what we had to accomplish. So that's, you know, that's both the good and the bad because the client asks for a, we bill a it's perfect. We test it, it's perfect. This is awesome. We build it. It's not what they want. We look at the requirements and I can't tell you the number of times I've been told. It's what I asked for, but not what I need. You gotta fix this and I don't have any more money. Each of those experience really good, really bad, really stupid are things that have informed my career and have been those pivot points to say, okay, here's the path we're gonna follow here. Here's the path we're gonna follow here. Does that answer your question?

ben parker:

Yeah, no, it's good. I think, obviously, I think everyone's. Your career shapes you. Doesn't it like anything can, challenging projects. You, you're learning constantly, aren't you? You, you gain that knowledge and then utilize it in wherever your career goes, in career, life, personal, whatever. I think it does. Everything adds to your ability to move forward. I completely agree.

tony giordano:

And one thing I would say, and this is for anybody who's listening, I've been successful and I've worked really, really hard. But my success isn't just it's hard work. It's always the element of luck. And here's another really important thing, and that is the people in your life. I have been super blessed, not blessed, super blessed to have some really good mentors, and I have them to this day. This, I'm a senior executive, but I have mentors that I will go to and say, what do you think of this? Or What would you do if you were me? And it's great to have that sounding board of people you respect. People are equally, if not more successful than you are. And say, what do you think? And they say, good plan. Here's your risk. Good plan. I wouldn't do it. Bad plan. Fix it this way. And my getting that great frank conversation from people you know and respect and can confide in is incredibly important. So my whole career, like I said, my, my first boss, James Martin, he's the guy that actually recommended me to go work at Oracle because you need to go work for Oracle. And um. He actually got me in, helped me get in, uh, introduced me to Larry Ellison. Larry actually gave me some advice. I don't, I doubt he remembers me because I was such a junior kid and he's met a million people since. But it's funny you meet people, you learn things from them and you incorporate that your corpus of experience and that becomes your sound internal sounding board to make decisions that you can be c in.

ben parker:

Yeah, totally. I mean, I think it's. You wanna just, whoever you may you wanna learn from, don't you wanna be sort of the, you don't wanna be the cleverest one in the room, sort of, it's you.'cause people, everyone's got special talents. You just. You can learn anything from anyone. Can't you really? If you look deeply, you can see what they're doing.

tony giordano:

So, Ben, the, the longer I'm in the business world, I don't wanna say older, but the longer I'm in the business world, the dumber I get when I say that, I realize if I surround myself with people smarter than me, we're gonna have the best art solutions solution in the world. Whether I'm building a client's thing, something for IBM, it's just having the right people with right eminence, the right skillset is the best thing you can do. So I just brought in a young kid yeah, I'll share his name. I, there's a lot of, his name is Usama lives in Morocco. He's probably 23, 24 years old. Wicked, wicked smart kid. He'll work 90 hours a day and you just look at him and you say, this guy's gonna be a leader. And he's only been part of my group for maybe two months. He's already done incredible amount of work and his opinion is already well valued. And our team, so the best piece of advice, give anybody is have good mentors and have good people around you both. Both from somebody to assist you, the mentors and people to work with you. On a daily basis. I know not everybody gets that opportunity. I, I've been, like I said, my career is filled with a lot of luck and a little bit of positioning, and then I have just enough sparks to make sure I surround myself with really smart people.

ben parker:

So then, have you ever faced a major career setback? I think we all do. Yep. You

tony giordano:

know, uh, I guess it wasn't a setback, but when I went to Merkel, I was planning to stay there first, a good long time. I wasn't planning to stay there less than two years. The owner of the company and I, there's some things I probably would've, I may not have taken that decision to join Merkel. I'm glad I did and actually I, it was a smart thing to do and the experience I got was really valuable around MarTech and real time and CMOs and business worlds actually smarter in data than oftentimes, uh, our, i, it brethren. But I, um. I didn't realize that this would be on an 18 month adventure, so I didn't really plan to have you look for something else at that juncture. And, uh, at the same time that happened. I also, little bout with cancer, so it was like two things hit me at once and I wasn't really prepared for, but you don't get the chance. Sometimes extenuating circumstances are things you can't control, so you just smile and say, okay, we'll deal with it. We dealt with both. I found myself back at IBM in a better position. And the experience I gathered in that two years in Merkel were valuable to this day. And there's, and again, another valuable set of experiences and people that I've met.

ben parker:

Amazing. Yeah, obviously it's, it's what you, everyone's gonna have problems. Everyone's got a setback. Um, everyone's gonna have personal problems. It's how you deal with it. Uh. You have to, I think it's obviously like time don't stop. You have to move forward. You have to just keep going, keep driving forward, and whatever the problem is you have in your life work time just does not stop. So

tony giordano:

for me, in my career, in my life, I've been doubly blessed.'cause I, I don't, I haven't got a job yet. I'm getting paid, but I'm getting paid for doing something. I might also, I might just do for fun on the weekends. What we do with data is so much fun. I've been telling anybody who'll listen, the world of data we're inhabiting now has so many more toys or so many more pieces of candy than it did five, six years ago. You know, many of us grew up in the data warehouse world. We're really a primarily one use case that was reporting in and, um, visualization. About 10 years ago, we had this thing called big data and data Lakes on DU really for data science purposes. Well, then we'd start doing digital. Then we start pulling all this stuff together in what we cultivate platforms and we had a, a common environment. Typically on a cloud or on one many clouds, where we had reporting and analytics where we had data sites and ml where we had digital, and now we're adding on Gen AI now a gentech with unstructured data. So the use cases for data and the types of data that we're using and adjusting and curating to do certain things is becoming so cool. And the tools that we're using are so phenomenal. That world of data has completely transformed in the last 24 months, and it seems like just in the last 24 months, it's, there's been two big, massive changes. The first one is Gen AI, the second is agen. And every time we think we've caught our breath, something else comes up. Say, okay, we can cut another 30% off our projects that we just cut 30% off before. The world we're in right now is in such flux that I can't think of a more exciting time to be in the data space and being in the consulting space because we get, we as consultants are usually two, two to three chapters ahead of our, our, our constituent organizations, at least a chapter ahead. We're getting to do some of the cool stuff out in the business in this space called data.

ben parker:

Yeah, and it's, I mean, you got AGI Other things coming up is who knows what's happening in the future as well. Is a lot's happening, isn't there?

tony giordano:

We are showing, proving at some of our client set the move from just even just traditional Gen AI, the quote unquote chat bot whole thing, but really creating agents to do things rather than just a Gen AI chat bot is gonna. Truly transform. Again, it's a second turn of the crank that clock. When I say change a clock, things that used to take a thousand hours, we've conservatively guessed can be 35 to 60%. We keep it at 35%'cause no one's gonna complain if you come in early. We don't wanna be too aggressive, but we know it's probably closer to 50. We also have to get people up to speed with some of the stuff, both the clients and the consultants. So, so even though it could be 50, we're, we're keeping it 35%. And now with, with AG agentic, it's an easy, now they're 20, 30% on top of that. So just think of it. Projects that used be like 10,000 hour projects are gonna be done in 5,000 hours or less. That's what's exciting about the tooling that's coming out and the processes. Now, by the way, we're still gonna have to build. A data model, whenever you build a data structure, whether it's, a object relational database in Snowflake, or you're building unstructured data stores in cloud object store, well, you don't, you gotta have a design pattern based on the different types of uses. Is this gonna be for transactional purposes, for normal form, is it gonna be for an enterprise data warehouse de normalize for, for second normal form for cloud out, cloud object, relational. Gotta have some sort of structure. But do I need to build a conceptual data model, which takes two to three weeks? Do I need to build a a logical data model which takes another three to four weeks? Or do I need to generate this, the structures which could take a week. Now we can do all this at once out of Angen tool and then have a human validated. By the way, you hear this term a lot. I'm sure you've heard it. Human in the loop. Well, that's what the human in the loop needs. We don't have to build the data model. We need to validate the data model that people are gonna build. We had a long conversation on that earlier today. Most of us grew up with parents who were exposed to this really cool invention called the calculator. And my dad used to tell me that there was stories where, you know, there was debates. Industry and in academia like, well, should we abandon teaching people math and just give'em the calculators or should we allow calculators in the school room? I mean, it was a huge, huge debate finally that, you know who can imagine now using even a calculator?'cause I don't know about you. I have this cool modern tool called Excel. Wait, no, I'm sorry. Excel is what? About 40 years old now? So we use tools that are now just commonplace and part of our everyday living. We just assume it, but we're still taught how to do two plus two and multiplication and division, all that. So while we have tools to do this excel and other artifacts we still need to learn math. We will have tools that will build data models, but we still need to understand what those artifacts are. Does that make sense?

ben parker:

Yeah. I think it's, like you said, we've got the tools. You still need that strategic thinking, don't you? Moving. Like I said, we've got, it's the technical tools. Now you can do the heavy lifting, you have to do coding as much. Now you just need, I think it's getting more business domain focus, isn't it? Around data is actually understanding the business problems and how can you actually create something that's gonna solve whatever problem you have

tony giordano:

but on. And that's, that's the most important thing is that we. Are gonna move into a time where organizations are gonna need data practitioners that understand the business domain I'm working in banking, retail banking, commercial banking, wealth investment management. Or I'm working in a telco and I need to know, have that industry expertise along with that domain expertise. Customer marketing is this around supply chain. So these are all things that, quite frankly, these are things we've gotten away from.'cause we have celebrated. Knowing Python and Scala and Snowflake and Redshift, and yeah, those tools are important, but the tools that use these technologies are becoming easier and easier to use. And it's less about defining the patterns into'em because they're already predefined and we know how to build patterns for third Noal form. Dimensional and, and no sequel. But what we don't have is people that know the domain and industry knowledge that can confirm what we generate out of these Gen AI tools is the right solution for that particular. Business problem.

ben parker:

Yeah. And then also you, you are getting, like you said, you touched on now then it's the tools there. Like a lot of people are trying to get more specialized in them when really, like, you should be focusing more, especially like early data scientists, like focus on the mathematical side. So many leaders say PE people are just not learning the, the, like they dunno understand the basics of mathematics statistics, which is, should be the. Foundations for your knowledge, and then you add the tools on top.

tony giordano:

You know, you're absolutely right. In fact, we brought up something a few minutes ago and I, I think it's a good time to celebrate it a little more. You know, some of the people who I, I was incredibly fortunate to work with in the beginning of my career, I brought up one James Martin. There's another guy, I, I don't know if you've ever heard of him. His name was John. Google him afterwards because, uh. He's the father of much of what we do with information as well. So James Martin and John Achman to the luminaries of the space. We, they built the foundation of much of the data work we do on to this day. He built something called the Achman framework, which is still brilliant, and the fact that looks at a system and looks at it from a, a data dimension, a process dimension, a network dimension, a people dimension, and. Conceptual, logical, physical substantiation. It, it was a very simple, easy pattern and I, I got a chance to meet him'cause, uh, his cohort,'cause John was semi-retired, actually was a neighbor of mine, and he actually became one of my mentors. John said, I, I observed the pattern. I didn't invent this. I, I was doing work, some work in, um, aerospace, and I just sat down one day and I, it came to me and goes, this is just a pattern that it's, uh, work. It's a series of rows and columns and he was right. And to this day, when we think of data it's a series of data structures. I don't really care how, what we call it, it hasn't changed and it's really a three-tiered model and every single data environment I've worked on from Teradata to Big Data to now ends up being a three-tiered architecture. That is some sort of data store where you're gonna put in raw data, some sort of data store that you're gonna conform. Some sort of data store for typical for the application you wanna use, what we call it is a data lake, a data warehouse and, and a consumption layer with data products in it. And, but it hasn't fundamentally changed in the 30 years that I've seen this business from college to now. It's very consistent and certain patterns will always be consistent the way you structure data for reporting the dimensional model. Something you know, Ralph in. I'm sorry Ralph Inman talk about it. Floridian Ralph Kimble did many years ago hasn't changed. Bill Inman's concept of, uh, the enterprise hasn't really changed. They're all evolved, but there's still important patterns that we have learned and you still need to learn the, understand these patterns to solve certain problems. But I can now instantiate those patterns in Gen AI to produce it. But then I gotta have the smarts to know, understand, okay, this is, this isn't gonna work. Even know Gen AI. Created this, it's gonna create a performance problem. The way we set up this fact and dimension data, you still need to know the math, but you don't have to. Do on a piece of paper. You don't need to do it on a calculator. You can do it in Excel. You can still do the model. You don't have to build it yourself in PowerPoint or Erwin. You can now do it in Gen AI tools. Ben, does that make sense to you?

ben parker:

Yeah. I think it's like, I think it's, again, it's, I have this conversation with a lot of leaders is you still, you need to keep the basics basically. Business is business like. So many people are like, again, in on the, like on the technical side, they're focusing on the talent tools instead of the like knowledge wherever your domain is. And then again, also like if you're talking, I was gonna dig into Gen AI in a bit, it's a lot of businesses are focusing on the Gen AI instead of actually the business. And so sometimes it's, I think people do get. They're chasing the shiny toy, aren't they in the market?

tony giordano:

Well, I think you just came up with something that I think everybody just saw the phenomen, and we do this with every new way, every the industry, they just run to it. And the research firms, the Gartner, the hss, and by the way, I have friends with these firms, but I, I do chime them for this'cause they, they do. They cause a lot of this stern, so do the software companies, the cloud companies and all of them. And that is we gotta do this. This is the hot new thing. And how many companies did all these POCs with, with Gen AI and said, okay, we got it to work but we're not sure what to do next. And a lot of those POCs were never equipped to production or they don't know what to do. Now some of it is'cause they don't really have a good data strategy. They start talking about their data quality, which I always find is pretty interesting because most Gen AI projects are gonna be done with unstructured data. Ben, I've been in the business a few years. How are you gonna identify good quality or bad quality from PowerPoints, PDFs, JP is beyond me, but the structured data problem that people have never solved. Yeah, that's still a problem but I. I think it's not just that, it's what are you trying to accomplish with Gen AI? What are you trying to automate to do better? Or what is the business outcome you're trying to achieve with this new tool that's gonna make your business better? You know, the thing I loved about digital, it was, we, it's creating new channels to interact with your customers and quite frankly, something that digital revolution is still going on. I don't think transformation is complete. We've kinda lost focus.'cause there's a shiny new object called Gen AI, which I still, I think is gonna evolve really. Digital transformation 2.0, which I think we're starting to already see. At least that's the phenomenon I'm seeing. It could be called something different. I could be wrong, but I don't think so. But it's the, the shiny object paradigm in, especially in it, it's the shiny new object. We gotta go chase it. We're not sure what we're gonna do with it. Once we do get an, we actually get a POC what now? So it gets back to your business strategy. What are you trying to accomplish? How are you trying to make your business better with whatever new phenomenon that's coming up? By the way I say phenomenon because you can put in digital, you can put in Gen AI, you can put in whatever, supply chain optimization, whatever, shiny, you know, whatever tool you wanna add. What are you trying to do with this new cool thing that'll make your business better? A lot of people don't go through that step.

ben parker:

Yeah. I think sometimes it's good just to keep things simple. When you try and make it complex with all these new shiny toys, sometimes you just need actually what, what we trying to achieve here. Just keep it nice and simple and then just keep it. But roll out from there.

tony giordano:

You know, one of the things, I'm actually a big advocate and look at when I was growing up and I was a pretty young guy and I was still at Oracle, I had a chance to work with, uh, talk about some luminaries. James Hammer and Michael Hammer and James Champion, the guys who wrote the, the business process re-engineering book, uh, reengineering the corporation, Oracle brought them in to, to give us some lessons and overviews and, some of their lessons really hit home. And I kind of put in my back pocket. I'm almost seeing, uh, hammer and Champion 2.0 with, with the Geek, you know, geek Way. I, I really love the Geek Way. I, I think that makes a lot of sense. One of the things I get from both those books, and it's really a simple but sage thing and we always tend to forget it about it, is let the perfect beauty, the enemy of the good. A lot of times people are trying to do things with technology, which are super fancy, super evolved, but doesn't really give a, the return on investment. It not, it doesn't give you any return, but you spend, you know,$2 million and you can now do a report three times faster than you could before, but it's really not changing your business. Well, it's not necessarily a good return on investment. It's what are you trying to do with the technology? Why? And make sure you got a proper IOI case to, to, or our oi case to, to put behind it. My, my, uh. MBA was an accounting Ben, so I was working at Pricewaterhouse at the time. I wanted to become a partner, not a principal. So I, I went through and took all the undergraduate accounting courses after my undergraduate computer science and accounting degree as a graduate. That's still keeps into my, uh, my point of view.

ben parker:

Okay. Let's, um, move on to the data topic then, because obviously Jen and I, I mean everyone's trying to do it. Should we say, so, I guess obviously key is, lending it into the business, isn't it Now, getting it into our processes. So with Jen, I, I mean, he's becoming the primary use case for many organiz uh, organizations now. And I guess, how is it reshaping business priorities? I

tony giordano:

think it's important to understand. I view it for my business. I use Gen AI both horizontally and vertically because. I run a business, I, I run a data consulting business, so I've used Gen AI to, to re-engineer how we do data from beginning to end, from strategy through delivery, through maintenance and evolution. So just observing how we've had to do it from our business has helped inform me how we help our clients do it. And, and it's, you need to really look at each of your different processes and say, are there opportunities to, to automate and integrate? More pervasively than you have in the past. So I think this is gonna, I kind of mentioned it a few minutes ago. I think this is gonna truly create, digital 2.0. How do we go through and do that next level of automation through all of the business processes that you have? I don't, by the way this phenomenon, I would be shocked if we're done with it in two years. I, I think it ha will not crest for at least three to five years. Could be wrong. But don't expect to see people really fully understand and exploit agentic AI in the next 24, you know, 36 months. It's just gonna take time for them to figure it out. Start evaluating where's your biggest pain point? Where can you start to automate to reduce your cost and time?

ben parker:

And yeah, I think it's. You touched it there. Businesses gotta find out their strategy, um, how they're gonna utilize it. Again, I think it's same with generative ai, isn't it? It's, I think it took, took businesses quite a while to get you to grips of what they're gonna do with it. So again, things like Gen AI, that's gonna be the same. It's not an overnight, alright, we're can do this. It's actually. That strategic thinking and how can we actually reshape this business to increase efficiency, grow all the business basics really, isn't it?

tony giordano:

You're spot on. You know, earlier this year, it was in April, IBM had a conference and we, we had a bunch of, uh, round table sessions and, um, clients from all over, world came to this conference, it's in Boston, and uh, we said, how are you guys applying gender to ai? And it, it was fascinating'cause I was a moderator and we had people. From IBM sitting, helping drive the conversations at like five, actually it was about eight different tables and about a third of them were doing some really cool things like, well, we already got this and this, we've already already got our next use case set up. Here's our problems or questions. How do we better set our data world to exploit this and some really interesting but technical, easily answerable technical questions. But then the other third were kind of like we, we haven't even started'cause running sure where to start'cause they really hadn't thought through. What are we gonna use a Gen AI, or I'm sorry, Gen AI four. And they, they were still struggling with what's the first use case and they're trying to find the perfect use case, by the way, that great example of organizations not being the perfect good, the enemy of the good, and they just really didn't have any strategy at all. And then there's people in that, in the continuum between them. But you could tell the companies who got it were already planning their third or fourth POCs. And also strategy, how to make this a. An ongoing organizational function are the companies that will probably be around in 10 years. Some of those who didn't get it, be curious to see where they're at on the stock exchanges in the next three to five years.'cause a lot of these companies just didn't get it, at least the people they had sent to this conference.

ben parker:

Brilliant. And yeah, it was sadly it would be, you think gonna be interesting. Well, it's interesting times ahead how business are gonna. Implement it, how they can adapt. Obviously you've got so many different businesses now and different journeys. Well, especially ones with a lot, the more mature businesses, obviously there's a lot more to unwrap, isn't there? And as opposed to the smaller companies, it's obviously a lot more, there's a lot of thinking to be done. Strategies need to be DA implemented and embedded with the business, so you're gonna fall behind'cause work efficiencies are gonna be improved by the firms that utilize it all.

tony giordano:

Well, here's the thing, Ben. None of this is, uh, none of these decisions and challenges are technical. What they are cultural. When I say cultural, we're not sure we wanna do this. And I've already seen that. We've had conversations with people like, well, you can, you can cut your time in half. We don't know if we wanna really do that. Why? Well, we have to employ a lot of people. Okay, then, why are we sitting here?'cause if you don't wanna save ti time and money, then there's. This may not be the right thing for you. We're starting to go against, you know, even in the consulting world, if I told I, I have told my colleagues at IBMI am not gonna need 7,000 data engineers going forward, I can do it with a lot less. Now that doesn't mean we need to fire 7,000 people or fire 6,000 to keep a thousand, but there's gonna be x number of thousand people that are gonna have to figure out what to do next. As consultants, and it's gonna be domain expertise. It's gonna be industry expertise, it's technical excellence. It's configuring, configuring these tools. So there there'll be new jobs, but the world's gonna change in a very radical way with this, with this drive, with the Gen AI and AgTech that I, I don't know, everybody's really ready for it. I guess it's the, the typical. Three-pronged way, which I mentioned a few minutes ago, and that is some people are gonna get it real quick. Some people are never gonna get it, and a big, big chunk of the world's gonna be right in the middle. So how good. Sorry, go on. I was gonna say, is it, do you observe the same thing?

ben parker:

Yeah, I mean, it's again, I mean I've had the conversation as well like about people that's come up, like, how are we gonna deal with that? It's come up from different managers. I guess again, also when we start speaking to like more the larger companies, it's, it's a lot of complexities there, I think, and also I think it's a lot of, it's a lot of knowledge gap that needs to be actually understood as well. I

tony giordano:

agree. I agree. Is it a knowledge gap or is it an acceptance gap?

ben parker:

Probably a bit of both. Uh, I think, I think also you, like you said, you, I think the fear element's gonna come into it, isn't it? Like you said, if you've got these tools, it is gonna take some jobs and then create new ones. Do people want to have them harsh conversations, isn't it? I think it's that cultural change is the challenging bit, I think for. Individuals.

tony giordano:

I mentioned earlier that, I'm a Midwestern kid and I grew up here in Detroit. My dad worked for gm and uh, I was a pretty young man, young boy. In fact, my dad was a lotto. He had me pretty late in life. He, uh, he and I were watching TV one night and there's a bunch of union guys for, I think Chrysler. It was really not around anymore, by the way, it now it's STIs. But they were saying, you know, we don't work for car companies. We work for the union. They were complaining about, automation and this, and they were gonna fight automations. It's like my dad said, these guys don't realize they're not gonna have their jobs in five, 10 years. Now. General Motors, when my dad worked there, had about 350,000 or 360,000 people. He was one of the biggest employers in the United States. It's not the biggest outside the federal government, and today it only has about 60,000 people. I think the same thing is gonna happen for a lot of businesses because of this Now, where all those people go, that, that's a political thing I don't wanna get into, and I, I. Wouldn't even know how to get into it. But I do know a agentic AI is gonna disrupt everything and everywhere in ways we haven't really imagined yet. I see it coming. I don't know the full ramifications of it. I know what it's gonna do in the data world. I've already seen it and we've already experienced it. We've got assets that do this disruption and we're continuing to them to evolve them that are becoming more and more impactful in terms of the. Amount of work, it's, it's, uh, disrupting and the amount of resources it's gonna disrupt, you know, in a positive way. In other words, instead of having five people, I need one person for two or three hours the benefits, the productivity of these assets and tools using AgTech, I don't think a lot of folks have have truly understood yet. So when it gets to, it gets both to acceptance and knowledge. So I think it's a little bit of both and it's just the sheer scope of the change that we're going through.

ben parker:

Yeah, and it's also, it's gonna be the reskilling and upskilling for individuals. You're gonna have to do it. It is just. There's gonna be massive change gonna happen. New jobs are gonna be created. What they are gonna be, we'll see in the future. But you've gotta adapt your skillset. Now. It's just, you can't just stay as you are. You need to constantly be learning. That's the way we are now.

tony giordano:

Tell you what we started this with, with me saying how much fun it is, being the data field. I'm gonna be Midwestern here. Ain't none of this is gonna happen without data. If you think of every agentic application, if you think about every Gen AI application, it's gonna require data. Now it's gonna require structured, unstructured transactional, semi-structured, but the need for data is to be. That fuel is gonna become more and more valuable and more and more needed as we go into these agent applications. It just is for those in the data space. Yep. You may not be doing Python anymore, but your skillsets are gonna be needed more than ever. So I know we started this with. The data world. We're gonna, I guess, finish it with the data world. But again, with all these broader changes, I just know the data field is gonna be robust and growing.

ben parker:

So how could companies effectively thread their Gen AI into existing processes without creating this chaos or redundancy? That's probably above my pay.

tony giordano:

How you do it without disruption, I don't think you can do it without disruption. I think. What you have to do is to take a safe space that is open to change and show how you do this. Then incrementally roll it out. If you try to do it, it's really the shock to the system because if you shock the system too much mm-hmm. You get resistance and will not be stopped because this will not stop. You may stop at one organization for a period then that that organization will fall behind. It's, it's inevitable. By the way, this is inevitable.'cause if one company stops it. That doesn't mean all everybody's gonna stop it everywhere. So that just puts that company at a competitive disadvantage. So how can you implement this new. Approach, I don't even wanna use the word technology. This new approach that will automate things in a way we've never imagined in such a way that will, not be as disruptive, minimally disruptive, and allow the disruption to occur organically from one part of the organization. That's a safe space to the broader part of the organization. That's the only way I see it. Really. Yeah. Going. In a thoughtful way because if you try to transform your whole organization, those things tend to fail. And then you're, then you're left behind by the people who do figure it out.

ben parker:

Yeah.'cause you've got, you've got the, I mean the change management piece is massive.'cause like you've touched on earlier, that cultural change, communication, it's, it takes time. It's not an like an overnight sensation. You need to just. Takes time to effectively manage that because obviously you've got you dealing with humans, different types of people, and that's not a fast activity. It's takes time to do it correctly as well. A hundred percent. So then, um, move to the next question. What does the evolution of data look like in a Gen AI driven environment? And then is it, is traditional data architecture still fit for purpose? Yes and no.

tony giordano:

It's a great question. I brought three, three third model. I'm still waiting to see, by the way, I, I've seen databases who had this in the past where you can actually embed true objects like PDFs and, and, uh. Video and all that kind of good stuff into an object relational database, but I, I haven't seen it at scale. I think the modern data architecture, especially the data lake, is a wonderful repository where you can start to gather unstructured data, semi-structured, structured that still goes into that conform layer. But try to conform data by customer for with bitmaps PDF files and, uh, JPEGs. Good luck with that and Ellen video and, and, and, uh, and, uh, Voix, I, I, I just don't see how that's gonna happen. So the traditional data lake is gonna be a good place to store that kind of data, especially. This stuff is big and it can get really big if you collect enough of it. And also pretty bulky, so you don't really want to keep it in high cost storage, like an object relational database, like a snowflake or a uh, any red redshift or. IBM, or I'm sorry, Microsoft Data Fabric I would keep it in Cloud Object store. So that's a good way to do it. The tools and how to ingest it and curate it are gonna change. Real time is is gonna change how we do both structured and semi-structured as well. And a lot of that's already in place. You know, I'm seeing more and more Kafka based architectures or API based architectures, and I'm seeing a little bit of both in most companies. So I think we're gonna talk about a natural evolution, but the one thing I'm seeing that Gene AI is starting to really promote is that collection for the first time at scale of unstructured data is now very focused. Where in the past it was a lot of science experiments, so the enterprise, uh, grade level architectures and approaches for pulling that data in for Gen AI, I think are here. Rag patterns enough, three years old. Most people figured it out or not, but those who have, are the ones who are doing some really cool stuff. I do find it interesting that you're seeing companies like JP Morgan Chase, which in my mind has truly pivoted to being a technology company. Building their own large language models kind of makes sense from certain perspectives. And I think you can see more and more of that as, as, uh, this whole industry continues to evolve. Definitely.

ben parker:

What, so what sort of skill sets do you think will become essential in the marketplace as Geni becomes more embedded in daily operations?

tony giordano:

Uh, I think it brought most of them up, I thought around domain. An architectural experience is gonna be really important. And that really does include both structured, semi-structured and, and unstructured data. That's gonna be really important that you understand those three domains.'cause it's not just structured data anymore. And how do you engineer it for performance And, uh, cost is really important. Like I just shared about where I would store my unstructured data. Mm-hmm. But then I've already kind of pivoted beyond that. Because here's the next big, uh, horizon. We're gonna be digging through these PDF files and these, uh, PowerPoint files and these scraps of paper that we have digitized and pulling information out, and we are gonna be correlating that and integrating it and aggregating it with structured data. As we start to do that now get back to more traditional data quality, but. Further down the the chain because you really can't do data quality on, unstructured data, but you can do it on the results of unstructured data that's integrated with structured data. So I think there's a whole nother level of learnings and tooling that we have started to think about and even started a solution, but that's gonna be part of the next phase of what we do with information.

ben parker:

I like that. Then do you do, I mean, do you feel companies are doing enough to prepare te prepare their teams for this gen off shift? No. A lot of

tony giordano:

folks are looking at it strictly from a, a technical perspective. I think they're looking at it, okay, we got this new technology. We gotta figure out what to do. I people wanna focus on this, this cool thing called rag patterns. Okay. That it's great. How are you chunking up your data? Cool. It's not, by the way, it's not as hard as it sounds. It does sound pretty cool. Like, Hey, I've, I've established rag patterns. Looks great on your resume, but it gets back to what are you gonna do with it? How are you gonna integrate this into the business? Now this is the same problem we a data science. 10, 15 years ago we were building these cool predictive models. I built some of those things. Here's my Merkel tree with my probability analysis on this particular chunk of your customer segment of people from the age 35 to 40 who lived in Manchester. I was doing some work in Liverpool. At the time. Okay, that's cool. Okay. What are you gonna do with it now? We're gonna, you know, make a decision on whether we should keep pursuing this or not. Like decision what on the project? No. Whatcha gonna do, whatcha gonna do with the information? You know, people would set up these data science experiments and they would. Evaluate results say, this is cool, but not really build it in their processes and make things better. That's the same thing I'm seeing a little bit with Gen AI. It's gotta go from the science experiment to the business value segment. So no companies aren't doing enough. They need to figure out how are they gonna set up teams in their organization on the business side, not in it. To solve those things with Gen AI and Gen Gentech and to grief. Let's just go back to the old predictive models. Whatcha are gonna do with all this stuff? Where are you gonna take advantage of improving your supply chain processes or your CRM processes or your financial closing processes with these tools?

ben parker:

Put it in, Tony. That's all the questions. And to be honest, I, I love your passion, your knowledge, and you've shared some amazing insights, uh, for the listeners. So thank you for joining the podcast,

tony giordano:

Ben. It was a pleasure and a privilege. Thanks so much and, uh, best wishes to you and your audience.