Summary
In this conversation, Kerry Boudreaux and Sanjay Venkatraman discuss the evolving landscape of AI, particularly focusing on generative AI and its implications for businesses. They explore the fundamental concepts of AI, its applications in organizations, and the importance of a solid data strategy for successful AI adoption. The discussion highlights various use cases of AI, the need for a structured approach to implementation, and the significance of agnostic solutions in technology. They also touch on the future of work in an AI-driven world, emphasizing the transformation of roles and skillsets.
Chapters
01:17 Understanding AI: The Basics and Beyond
04:10 AI in Organizations: Use Cases and Implementation
08:50 FourthSquare’s AI Solutions
11:50 Building a Roadmap for AI Integration
18:08 The Future of AI: Transforming Work and Skillsets
22:35 Conclusion
Kerry Boudreaux
I’m Kerry Boudreaux, Senior Vice President of Sales here at FourthSquare. And I’m very excited about today’s topic, Data Analytics and AI, the FourthSquare approach. And I’m even more excited about the guest that we have. He’s our Senior Vice President of Data and Data Analytics Practice, Sanjay Venkatraman. Sanjay, welcome.
Sanjay Venkatraman
Hey, thanks, Kerry. I’m super excited to be here and I will do my best at explaining AI.
Kerry Boudreaux
Awesome. Let’s dive right into it Sanjay. There’s a lot of talk around AI these days. Can you share with us your version, your explanation of exactly what is AI?
Understanding AI: The Basics and Beyond
Sanjay Venkatraman
Sure. AI has many different facets, but I’m going to start off with what it is at its very core. At its very basic core, AI is like a smart helper that we teach by giving it some examples. And it uses those. It assimilates, it learns, and it’s able to help us with a variety of things. Simple tasks like answering questions or recognizing persons in a picture. And in more complex scenarios, even driving cars. But even though AI can be really smart, I want to emphasize that it needs us humans to teach it and tell it what to do. And there’s just a lot of variations. There’s complexities, there’s simple versions of it, but at its core, that’s all it is.
Kerry Boudreaux
Exactly. Would it be fair to say that AI has been around for many, years, but it’s this new emerging technology called generative AI that is brand new? Would that be a fair statement?
Sanjay Venkatraman
Yes. AI has been around for a long time, things like image processing, things like recording pictures. These things have been there for a very long time, but what is new and what exploded on the scene with technologies like ChatGPT really is having a technology that can understand human language and be able to generate a cohesive response back. So that’s new.
And because that’s come into play now over the last couple of years, it is exciting because it helps us solve a variety of problems which we could not otherwise solve with just technology alone.
Kerry Boudreaux
Yeah, so I was recently playing golf with some folks in the music industry. And I asked them about AI and they were super scared because of this new generative AI. It can generate your own music. There’s no creativity when it comes to the musician. And you talk about it within the film industry and they talk about their fears of AI.
You can even talk in the educational world, in the classroom, about how ChatGPT can be used to generate your own research. You don’t have to do any research. Just tell ChatGPT to produce a research paper or a book report, and it will do it. So, I understand that there’s fear around AI.
AI in Organizations: Use Cases and Implementation
Kerry Boudreaux
Within an organization, what are some of the use cases? In my opinion, we shouldn’t be so fearful of AI, because I see a lot of excellent use cases within business applications and organizations. What do organizations need to know when it comes to implementing this next phase of AI, this generative AI?
Sanjay Venkatraman
Sure. So I want to take a step back here, just talk through some challenges that businesses face. And I don’t want to jump into AI because many times, what I have learned talking to customers or prospects, leads, et cetera, is that many of them don’t know what AI can do.
I think the first approach really is understand their problems, understand where they’re struggling and then give them, provide them solutions. Most of them AI, some of them could be traditional solutions too. When I talk to leaders and VPs and CIOs, et cetera, I sort of classify their problems maybe into three or four different types.
The first one probably is automation. They’ve got individuals that are just doing wrote work. The prior technologies that are available are not suitable for their kind of automation. So, they are just doing stuff manually every single day. So I see that as a huge opportunity for AI and as a problem that businesses face.
The second one that I see most often is leveraging AI for decision-making. And what I mean here is that organizations today have just a data overload, not an information overload, a data overload. We have individuals in organizations that can run reports, that can look at the past, but how do you make sense of just the gazillion amount of data that’s generated every day within and outside the organization and make sense of it? So that’s another huge area of opportunity for AI and challenges that companies have today. So it’s like a match made in heaven for AI.
The other two opportunities that I also see talking to senior executives is around retrieving the right kind of information at the right time. And I don’t mean structured data. I’m talking data about the company. It could be things such as data around contracts, around purchase orders, around policies and procedures for HR, making it easy for people to both access and digest that information. So I see that as a huge challenge.
Following that, one of the big challenges of today is not just the one time hire, but how do you make them productive within an organization and just reduce the time it takes for somebody to get on-boarded to the time they become most productive and efficient? That’s a huge challenge for companies, especially post COVID where we don’t see each other every day. So that’s become a huge challenge and AI has a number of ways of doing that.
Then the final thing I’d say, or add on is, optimizing business processes. This is a sweet spot for us really, this kind of segue into what we do really well is, and this can be for small organizations, large organizations, where companies have grown, they haven’t really spent the time to go back and see what is it that we can take away or can we optimize their process?
They haven’t done that work, so they’re left with suboptimal processes. And then for people, things keep getting added on, right? Just stacked on more and more and more without looking at the foundation to see are there ways to optimize it. AI is a great candidate for doing some of this work. So I see all of these challenges as I sort of talk to companies and AI has technologies that we can put forth to help solve some of those problems.
Kerry Boudreaux
Awesome. Yeah, so when you were talking there, I see two themes here. I see AI improving productivity, and I see AI improving work quality. Would that be fair to say that those are the two of the biggest benefits that come out of AI?
Sanjay Venkatraman
Yeah, I would say yes. And there’s the creativity aspect too. And that comes next from generative AI. Yes.
FourthSquare’s AI Solutions
Kerry Boudreaux
You and I had the pleasure of working on a recent project, working with a public-sector agency. We won’t name the agency, but tell me about that project where they had the city records archived and they had this big warehouse full of these records and they were all paper and what exactly did they want us to do and what did we propose and the benefits that we that they saw the benefits that were going to be gained from this.
Sanjay Venkatraman
Sure. before I answer that question here, I want to at least introduce AI and what we have at FourthSquare, because it fits really well with an AI offering that we have. So just talking through some of those problems, what we’ve done over the last six or seven months is develop three offerings for customers. And one of the offerings is our content hub, or AI Content Hub.
AI Content Hub is able to take pieces of paper or documents or pictures and is able to digitize them and then extract useful pieces of information from those non-digital documents, so to speak. And that’s something that it does really efficiently, it does cheaply, and there are technologies that we can put up really quickly.
So with that in mind, we took that approach to this public sector organization. They had records going back to 1890 that was in paper stored away. The goal of this exercise was to take those pieces of paper and then look at it time-wise, to go in reverse chronological order and start scanning those pieces of paper, converting that to possible PDFs.
Then using our content management hub to extract useful data from it, they’d be able to index and store the data in a digital repository that made it easy for somebody to ask a question and say, hey, can you tell me how many building permits were issued in 1898? Or what were the building permits issued for this particular street in 1890? And voila, we can pull it out in a matter of seconds.
Currently, it takes like hours because it’s actually stored physically somewhere. They’ve got to go in there, look things up, and if they’re lucky, find it. So we have reduced the time it takes and the manpower it takes to search for documents and provide that information over to residents of the town.
Building a Roadmap for AI Integration
Kerry Boudreaux
Awesome. So if an organization wants to leverage AI, how do they get started? What does that process look like, Sanjay?
Sanjay Venkatraman
Sure, that’s a great question, Kerry. I think companies struggle because they hear a lot in the media about AI. They’ve seen it do some really advanced things. Some of that is pure hype. You can’t do it today, but they hear about a lot of things.
Where we usually start is just talking to organizations, non-AI, like I said. Talking to them about what are their challenges? What are these struggles? Where do they see themselves in three or five years? That sets the stage for what, three or five years from now, we can kind of anticipate sort of the problems they might have, and then their current struggles. It could be with productivity, it could be with people, it could be with other things. So we take that all into consideration and then provide them with a step-by-step approach for AI adoption.
That’s what we do really well. And that step-by-step approach is really laying the groundwork for a three-year plan, but not overwhelming them with all the details. We say, hey, the first six to seven months, take baby steps. The crawl, walk, run approach. Take these baby steps. Take this first step of, and usually that comes under either productivity improvement, automation, or with information retrieval.
Those are the projects that are usually low-hanging fruit, where people can understand things and the greatest return on investment. We start there, however, we tell them, here’s your three-year roadmap to become sort of future-proof.
Kerry Boudreaux
Yeah, you talk about that three-year roadmap and I think of the term strategy. You you really got to have a strategy in place and then I think of your years of experience and IT leadership. Would you say that before you do anything, before you buy any products, whether it be, you know, Snowflake or Databricks or buying any of our solutions, AI solutions here at Foursquare. Would you say that you really need to sit down and come up with an overall data strategy and what does that look like?
Sanjay Venkatraman
Sure. So a couple of things that I’d like to add here is there is, I’d say, a misnomer in the industry that you need great data to implement AI. On the data side of things where I talked about how companies are inundated with data, they need to make sense of it. Yes, you need good data to leverage AI.
But then when you talk about productivity improvements and then talk about how we can optimize business processes, those don’t rely on structured data. Those rely on process data and definitions, which is still data, but it’s likely different. The first step is separate the two out.
It is important to understand what AI can do, but AI is not a replacement for Snowflake or Databricks. You still need those tools to be able to efficiently model what the data is going to look like, extract and transform the data, and provide the data in a repository that AI can then act on. Those tools are still needed. Those processes are still needed.
Once you have that in place, data that’s structured, then comes the turn of AI. Then we create models out of that. Usually they are predictive models. Usually it’s a natural language models that we develop on top of what Databricks and Snowflake have already done. So I’d say it’s sequential. AI is not a replacement, but you need good data. You need a successful Databricks and a Snowflake project to even kickstart an AI initiative.
Kerry Boudreaux
Awesome. Coming up with that overall data strategy is really key. And that’s something that you can do right here at FourthSquare, right, Sanjay?
Sanjay Venkatraman
Absolutely. As I sort of lead the practice on both data and AI, we look at that holistic evolution of an organization, we absolutely look at data first. So we look at data first, how do we extract data from this multitude of systems? We work with the senior leadership of organizations to understand, you know, their growth plans, their KPIs.
How do they measure success of their individual departments? And then based off of that, we create KPIs and then drill down those KPIs into metrics and into data points. Then we attack all of these source systems to make sure that all of those data points are grabbed and then they’re standardized in order to make those KPIs a reality. So that’s what we do on the data side of things, Kerry.
Then once the data side is done, the AI side kicks in and says, hey, we’re done. We’ve given you these metrics. Now, can I predict based on your growth plan? What might happen in three years or four years? What might happen if I introduce this new product? What might happen if we acquire a new organization in this category? All of those things require good data. And then once that’s in place with our data projects, then we can have our analytics and AI projects on top of that. So as a service, we offer like that full service from data to AI.
They can go from having almost no formal reporting to becoming a specialized organization in the AI realm.
Future of AI: Transforming Work and Skillsets
Kerry Boudreaux
So one of the things that you and I have talked about and I’ve actually heard you with customers and it really resonates is that the data tool sets that we have here at FourthSquare are agnostic. I mean you don’t need a particular platform, whether it’s Azure or Oracle or, whatever solutions can lay over any platform. Can you share a little bit more about what that means to be agnostic?
Sanjay Venkatraman
Sure. So as a practice, our practitioners are technologists, but they’re not technologists first. They’re practitioners first. What do I mean by that? They understand industry verticals and they understand the process that it takes to get to a final destination. That could be structured data in some form or fashion.
Once we understand the process, because we have the technology expertise, we can then come in and say, you know, for what you’re trying to do, organization A, we think that Snowflake is the right tool, or we think that Fabric is the right tool. We’re not a technology first company. In other words, we don’t go in and say, we’re a Microsoft Fabric company, and then limit our scope to the capabilities of Fabric only.
Kerry Boudreaux
Awesome. So we’re getting close to the end of our episode. And is there anything else that you would like to share in regards to what we do here at FourthSquare in regards to data and data analytics and AI, or anything that burning question that maybe a CIO has in regards to data and in particular, AI?
Sanjay Venkatraman
Sure. I think it’s important to talk to customers about their challenges. Don’t go in and say, we’re like an AI company as in sales. Understand their challenges, then propose solutions. And in many cases, like you rightly said, it could be, we’ve got to approach, have a data first approach. We got to make sure that data house is in order before doing AI. Go in with an understanding, an open mind and understanding that it may not be AI first, given where the customer is at.
So, and that’s a big part of this. But having said that, I think that the next evolution of technology and business process really is going to be AI-driven. What I want to add is a couple of things that you’re going to hear over and over again. While AI is transforming tasks, it is also creating new roles and redefining skillsets.
The focus I think will increasingly shift to more of a hybrid work. I don’t mean hybrid in the sense of COVID, but hybrid in the sense of employees working with their own personal AI agents. So you are going to enter into this world in a year or two, it’s coming, where you’re going to have your own personal AI agent, where you can offload a ton of your work. You can get to do some work, but you can also offload that with an understanding that tool set is going to know not only the technologies, but also the process and how to execute that process. And finally, for businesses adopting AI, it’s not just about adopting new tools, right? As you think about this new hybrid AI approach, you have to rethink workflows. You have to think about how staff can be retrained on AI capabilities, finally building a culture that embraces this technology.
Whether we like it or not, it’s coming and it’s coming fast. So I’d say from where we are, yes, we have a whole breadth of what we can solve for. But I think having the end in goal, we define how organizations can get through that whole process and be successful at the end.
Conclusion
Kerry Boudreaux
Awesome, Sanjay. Listen, I really appreciate you joining us today. This has been a fantastic conversation. If anybody out there listening to this episode, if you want to learn more about our data and data analytics and AI practice here at FourthSquare. You can reach out to me, you can reach out to Sanjay on LinkedIn. You can go to our website at Fourthsquare.com. We have a forms page, a contact page where you can fill that out. We will get right back to you.
I’ve been on discovery calls with Sanjay and it is fantastic to hear him engage IT leadership because Sanjay has been in your shoes. He’s been an IT leader within organizations. Thank you again for joining us here on the FourthSquare Solutions Podcast and until next time have a great day and good luck with leveraging your data in your organization.
If you’d like to learn more about FourthSquare and the services we offer, click here to contact us. You can also call us at (972) 919-6135. We’ll be happy to speak with you.