PegaWorld 2025: Transforming Legacy Systems, Fast: A Journey at the State of New Jersey

PegaWorld 2025: Transforming Legacy Systems, Fast: A Journey at the State of New Jersey

Hello, everyone. All right. Cool. We're on. We're live. The doors are closed. Welcome. Thanks for joining. Um, we got a great session for you today. It's the last one of the day, and I'm really happy to be with everyone here. So why don't we get started?

Uh, so I'm going to introduce the folks here from new Jersey courts. Excellent client people we love. We're very, very glad to have them here today to talk about their story, transforming their legacy systems and working with Pega to to modernize and to get off of their old mainframes and legacy systems and move into the future. My name is Sean Callahan. I'm a product marketing manager with Pega. I'm joined by Lakshmi Narasimha murthy and Edwin Fernandez. Thank you very much for that pronunciation everyone. I've worked really hard on that and I'm just going to start with a couple quick words to get things going here, get the juices flowing, and then I'm going to pass it off to these lovely folks who are, I think, who you guys are most interested to hear from today.

But let's start out with an analogy. Love. Good analogy. So I want to ask folks here today who wants to take a voyage with me on this beautiful vessel called the Titanic? Anybody want to go on the Titanic with me? Any takers? The lights are bright. I don't think I see any hands. A couple hands, huh? I have a movie I can show you. And it might change your mind. About what? The voyage you'd like to take.

So look, when I think about the companies that we work with, enterprise clients who want to transform, modernize, move off of legacy systems, it can feel a little bit like you're going to take a ride on the Titanic, right? It is an expensive Journey. Pretty penny for one of those tickets. Okay. It takes a long, long time. You might not end up where you're going to get to, right? And it is full of obstacles. There are icebergs everywhere, as you saw this morning.

And as you've maybe have seen in some other keynotes and sessions and throughout the booths, we're focused on legacy transformation because AI has made it possible to make this much, much easier. And so for our approach that we're going through now with Pega is to analyze your legacy assets, analyze the current state using AI, reimagine with Blueprint, and then deploy to the cloud for a future proof architecture. And we have technology that makes that a lot more possible than it used to be.

But it's not all about technology. Transformation is a lot about change management. It's a lot about people. It's about continuous improvement mindsets having the right champions right, having the right strategic approach because you can have all the tech in the world, but if you don't have those other elements lined up, you might not still get to where you're going to.

And that's why I'm really thrilled that new Jersey courts is is joining us today, because they have exemplified what it takes to transform. You need to have a commitment to technology and tools, but you also need to have a commitment to people, to strategy, to know what your goals are and to and to take an approach that lets you step by step and consistently move towards those goals. With the right champions and the tech kind of supporting everything underneath.

So that's enough from me. I want to hand this off to Edwin and Lakshmi, and I want you guys to all give a very warm PegaWorld welcome to Edwin and Lakshmi to take us on the rest here. Thank you.

So that's just doing this. All right. So welcome everyone. Good afternoon or good evening. If you're back in Jersey, um, I'll talk a little bit about today's challenges and what we're facing. Right. Um, our business, our like, a lot of big state or big companies. Uh, we run our critical workloads on the mainframe. That's where the bread and butter is. That's where all the magic is happening. Um. And no one knows how the magic is happening. That's the big problem, right?

So we are embarking on this journey, along with Pega and others, on trying to extract those business rules, trying to extract all of the information from what is currently there that someone wrote 40 years ago and no one still understands. Right. I would say that we are in our fifth five year plan, right? To modernize ourselves out of the mainframe into new technology, uh, into new horizons. Uh, we're now estimating, I think it's going to take the next 7 to 10 years to really get off the mainframe, take that workload, understand those workloads, and transition them not only to some of the Pega workloads, but also to other environments as not all workloads. You can't use one tool for everything.

Um, some of our goals, right? Customer experience. It's important, and it's imperative that our end users, law enforcement officers, attorneys, judges, uh self-represented litigants, um, and just normal constituents, if if you're anyone in new Jersey some way, shape or form, you're having to interact with the state of new Jersey courts, right? Uh, whether it's for jury duty, uh, or you get a traffic violation because you're speeding, uh, whether it's because, as Lakshmi pointed out earlier, you know, maybe you took out your garbage and left it out on the wrong day or something. Um, your interaction with the court matters. So the customer experience is important for us.

Operational efficiency. It's also very important. Obviously you want to do more with less. You want to have efficient workflows. You don't want something that can be done in one day. Take five days. You know, government red tape, things people sometimes think about. We want to have workflows that are efficient and get to a resolution as quickly as possible.

A culture of innovation. This is something that Lakshmi and her team play a pivotal role for us. Um, we have to welcome new technologies. We have to welcome and embrace change. We cannot implement the old way of thinking to address new problems of today, right? And Lakshmi team is instrumental with our enterprise architecture group, uh, to bring new ideas to the table, challenge our developers, challenge the business, challenge the stakeholders. Um, and that innovation is important. And we're seeing a lot of that today with artificial intelligence and many things to come. Our partnership with Pega has really helped us in this area. Right. Pega has been tremendous. Um, their their customer management team, customer success team, they've been, um, available to us all day, every day. Uh, they already know that with us, facts don't have feelings, right? Because we're brutally honest.

Um, so moving on to the next things. Uh, part of, uh, of, of our success is to have fewer silos, right? When you look at the court, you have criminal court, civil court, family court, municipal court, probation. There's all these different domains. But at the end of the day, they process case management very similar. Right? But they all have their own intricacies. Um, And having those reusable assets and eliminating some of those silos that we have today is paramount. And with that, we would accelerate our product development by introducing platform engineering teams that build reusable assets. Reusable components will drive our time to market. Next.

Okay. All right. So this is somewhat of a timeline of our journey with Pega. Um, we started in 2012. We started with Municipal Inquiry, which is basically looking up information on, you know, traffic offenses or municipal complaints. Um, in 2012 and, uh, we went from there into, I would say 2015, give or take, uh, criminal justice reform was initiated. Legislative mandate. Um, we have to do. To a complete new business process for criminal justice reform.

Just a short introduction on criminal justice reform is to reduce the people that are incarcerated because they are not able to pay any bail amount. So new Jersey started this program, wherein it would look into the risk factor of a person based on their history and then, um, decide if that person should be in jail or should be released pending trial. So we were at a critical point where we were not sure whether we should do Pega or we should go. Because before that we had a lot of Java applications. And in 2012, that was the first and only Pega application we had.

So we have an audience, our CIO, who was a project manager at that time, he made the critical decision to build a bunch of our criminal application. Bunch of the criminal justice reform applications in Pega, and that turned out to be a turning point. The most critical one of that is the public safety assessment tool, which is very complex. And it's it predicts whether the person is likely to commit a crime based on their past history. So he made the decision to build that application in Pega. And, uh, the rest is history, right?

And it was perfect for us to move to the Pega Platform because everything was very workflow based. Um, the business rules engine, again, Pega bread and butter really helped us, uh, implement the algorithm for the public safety assessment as well as all of the other components in the criminal justice reform. From the moment someone is arrested and they're fingerprinted right away, we're running a public safety assessment. From the moment that they have their first appearance, they're entitled to a speedy trial, right? They go through, um, Pretrial services program to determine, uh, you know, a few things there. They also go through pretrial monitoring if they're released. Right. Um, so all of these areas are critical and important, and they all have their specific workflows. And very complex workflows. We were able to successfully build them in Pega, I think version seven dot x. Yeah.

Um, from there. Right. We took that those lessons learned. We did all of that with Pega seven X on prem. Uh, our next journey was to migrate to the cloud to help us with our disaster recovery, resiliency, fault tolerance and everything else. And that has been a tremendous success as well with migrating to the Pega cloud. Migrating, uh, the, the operational, uh, portions of the administration of patches and all of that that previously used to be our middleware team. Now Pega is taking care of that, and we focused on what's important, which is developing our applications. We are now in Pega 24 .1.2. And soon going to Cloud K Cloud three whatever branding Pega uses.

Um, as you see, we're continuing to mature in our journey with Pega. Uh, once we went to the To the Cloud, we decided to establish a cloud center of excellence. Um, and enterprise Architecture and Infrastructure team. And the sole purpose of that was to have a co around Pega to make sure that we're following the guardrails, that we have the best practices, processes. Uh, what's new from, from, from Pega leveraging them. And I'll let Lakshmi, uh, fill in all the gaps.

So what started was, um, when Pega was introduced to us, they talked about Pega express methodology. So our CIO, who was the project manager at that time, um, he took that methodology and he customized it for new Jersey courts and came up with a checklist and we called it the UCM checklist. It essentially lists all the things that a project should go through to ensure that it's successful and deployed in production with good quality and all. So, um, that in turn gave way to, uh, the gate reviews and other things that we do with the Center of Excellence.

Um, but as you can see, today we are, uh, 90 plus Pega apps. So whenever I'm talking to people and then like how many applications, they are astounded by the number. So, um, but it's been a good journey. It's certainly been a good journey. And Lakshmi is going to get a little more into those gate reviews that we do as well, from the Center of Excellence in a slide down.

Um, but the the future, what does that hold? As you saw in the keynote, Alan's talking about legacy transformation using Pega. And this is where I think Blueprint is going to help us move forward in that journey. Reduce the time to market for our releases, our products, and really have that trustworthy AI with Pega. So, um, I'll hand it off to Lakshmi to go over the next.

So as you can see, with 90 plus applications, we have a lot of teams and a lot of developers, a lot of lsas. Uh, so we have this process of, uh, we have a separate business team. They are here to, um, so what we do is, uh, the first stage is at the ideation when somebody decides this is likely to become a project, we they meet with the center of excellence, and we also include all the other Lsas and all the other teams so that, uh, we they have to produce a checklist.

The checklist is to the idea behind the checklist is to make sure the business has thought through the basic things. Uh, what is the journey? Who are the personas? Um, do we have any integrations. Have we thought through the integration? What is the expected volume? Have we thought through those things? So we have the business present the checklist where everybody else listens to it and they ask questions. We are a very open organization where we are allowed to ask questions and, uh, and expect answers. So that's in the ideation stage.

And then uh, the next stage is the once the ideation is ready now, now they're ready to build the application. We have the gate to review. Now this is focused more on the technical side. We ask the Lsas to produce a design document. They have to list all the case types. Um they have to list all the frameworks that they are going to use. They have to list what integrations. Um, it's pretty technical. They have to, um, present. And again, uh, it is not just presented to the team, it's presented to the entire technical team where all the lsas and supervisors and managers, they all are free to attend and they can question.

So this is this helps because many times we get new people and they might not know that an integration existed. Or there's already a data model, the property that they could use. We also have the concept of EDC which is enterprise data pages. So we so we make sure that that way we promote reuse. So we make sure that they are identifying if they don't know. We tell them that you have to use this. So that's what happens in gate two.

Then after that the application once it's blessed, the design is blessed. It goes through development. And usually we ask them to present something at least every 1 or 2 sprints. So they go through they, they they give a playback of what they've developed. And then we ask questions and then we ask. They do a playback. We monitor the PDC for any alerts and then we ask questions. So that way we are ensuring that through every step we have a solid application that's being built and it can coexist with all the other applications. So our infrastructure is not in jeopardy.

So then we have the pre-launch, um, which is just before going to production again, they have to present the application to us. And then we ask them, have our security team runs scans for vulnerabilities. So they have to finish the security scan. They have to I cannot read from here, but there's. Yeah, there's security scans. There's there's post-production. Pre-production, a lot of pre-production tasks that they have to do. Uh, we also have a code review template. We make sure that they are filling the code review template, and then code reviews are done and the best practices are followed, and then the application goes to production. Without the sign off of I, the application doesn't go to production.

We also have taken turns. It's not that I is the only traffic cop we ask. We take turns from other lzs. They join the gate reviews for applications that are not theirs, and then they also, um, present or they also run the review sessions.

In addition to that, we also have design clinics. Let's say some application has a unique requirement that they have to use big extract or something different that they that we've not done before. We have design clinics where all the lsas and anybody that's technical in Pega participate. And it's not just we don't just go for Pega solutions. Sometimes we might say, oh, maybe there's an AWS solution. An example was AWS pinpoint to send text messages, text messages and voice calls. So we used. Then we said, oh maybe that's a better solution. So we have design clinics where the entire technical team comes together and uh, we ask them questions. And that's how we ensure that all of our 90 plus applications are very critical. Uh, because once when a person comes to the code, they are very stressed. We want to make sure the application is working. The code staff is able to use the applications and, um, the customer experience that Edwin talked about. So we are very particular about those things.

Now, I don't know if anyone here has gone to Catholic school, but um, everything Lakshmi was saying is basically like a sister walking around with a ruler, if anyone knows what that is, right? So the asking of the questions. I don't have a ruler. He has a yardstick. So that's that's how the process goes. Again, at the end of the day, what you want is to have quality product at the end of the day, and to have quality product, you have to have discipline. You have to have a process. And we've established that and that's working well for us.

Here are some of the applications that we thought we wanted to highlight. Um, everybody that's a citizen has been through the jury process. You know, when you get the jury summons, we all start thinking of excuses that the judge might approve. But we built previously we had the jury system in a power builder. Very old system. There was only exactly one person that knew how to, um, how to manage that, uh, product. So we rewrote it in Pega. It was very successful and well received. And it works, uh, really well. We did such a great job with jury that the one person that was supporting it retired. So we're very happy with the work that we did there. Again, speaks to the legacy moving to new technology. And when you look at other Um, things that are very important to to the public, such as Ada compliance and things like that. These are all things that in, uh, enterprise architecture, we're going to be plugging in and saying right now as part of our gate reviews is, are you building Ada compliant screens? Are you doing this, that and the other? So it's evolving process, right? So every day we keep adding new things.

Uh, the next application is the municipal complaint system. Like you said, your neighbor's dog is barking consistently and it's like you're not able to sleep. Next day morning, you march to your, um, police station and you lodge a complaint that gets entered into the municipal complaint system. Or it could be you witnessed a crime, or there is an arrest or a drunk right down, down the shore. Weekends. Um, there's a lot of entries made because there are a lot of young people at the show down the down our shore and. Peak season right now in new Jersey. So that's a very complex system, uh, that, that we built in Pega and currently we are working on the enterprise warrant system. Edwin talked about how we have a siloed system. We have a mainframe system for criminal. We have a mainframe system for municipal. We have mainframe system for family. Warrants can be issued in any of the systems. Um, so we are we what happens is sometimes because the systems are siloed, the court staff might forget to, um, nullify a warrant in one system. So when a person is there, they let's say they have a warrant and two systems, but it should be nullified in both the systems. In one system, it is another. It is not when a cop catches them for some traffic violation and they run, it would show up as though they have an active warrant in one of the system, which is a big loss of liberty, a person is arrested unnecessarily. So that's a big problem.

So we wanted to avoid that. So we um, that's one of the reasons why we built this enterprise. We are in the process of still building this enterprise warrant system that integrates across all the silos we have, and it builds a unified system. So there is only one place you have to make a change and it gets notified, and all the others updated in all the systems. And it's going to be a big, big win for the state of new Jersey. And anybody that has a warrant in the state of new Jersey. So Pega has enabled us to build really good systems. And these are just a few of the just three of the 90 applications.

So, um, at this time, I would like to, uh, give a big shout out to our Customer Success team. They are extremely helpful. Uh, they, we come up with all these innovation problems, new problems that we have not done, and then they know exactly who to get for us in the, uh, Pega side. They've also given us a team of technical people. So whenever we have questions that we are not able to solve. We ask them, for example, we wanted to do event driven books extracts. So then they put in touch with the customer, with the technical team, and they showed us how, uh, in the latest version of I think it's 24.2 upwards, you have event driven books extracts. So they really help us. So, um. Thank you.

Yeah. All right. So our environment is not solely Pega, right. Obviously if you see on the left hand side, it's a big mainframe. That's where our critical workloads run. But we also have IBM workloads, AWS Azure Java workloads that run on prem and some that are going to be running on the cloud. So we have a mixture of technologies. And one of the benefits of this mixture of technology and working with Pega is the integrations and all the connectors that we've been able to leverage and use to pull data, push data and integrate with the different systems. So we're looking at the future. Yep.

So moving forward again. Alan touched on this, right? AI is a big part. It's going to be, uh, even if it's not AI products or ingesting AI in their marketing. And we see it as two ways, right? There's the human led AI with Blueprint, which we've done some great work with Pega with their lighthouse team, which we'll get into. And AI led source code analysis, which is your looking at your code, looking at your COBOL programs or looking at, uh, Java code, etc.. I'll speak to the human led portion and I'm just going to speak on the AI led source code.

So on the human, we initiated a project in new Jersey courts where we took a mainframe system. We brought in the subject matter experts, the business, the stakeholders, and we basically had, you know, uh, all day meetings for two days. And we said, what is the system? Who, what, where, when and why? Right? How does it work? Why does it work? What are the processes, the flows, resolutions, etc.? And we started documenting all of these things just so that we can take our smallest mainframe system and discuss how how will it work if we leverage AI and leverage Blueprint?

What came out of that was one big prompt that says, this is how the system works, again, generated by humans, by meetings, meeting minutes, user stories, etc. we pop that into Blueprint and out came out a application, right? Um, we see a lot of promise with that approach. Taking plain English. Here's what the system needs to do and give me something. And the technology so far has evolved to a point where when we started this, it doesn't have as much as what it has today, right where you can now upload DLLs, you can upload a lot of information into Blueprint that will help it create a better output for you. So what we've seen so far with Blueprint is, you know, very, very rewarding. And the future is bright. Promising. Promising. Yes.

So, um, that's one of the things, right? We talked about constantly improving our gate. So Edwin and I were talking this morning about, uh, improving our gate review checklist, requiring every application to at least go through Blueprint. We may not exactly take what it is, but at least go through Blueprint see what case types it's given, what personas, what data model, even if it can integrate with our existing data model. So see all those things. So we are thinking of, um, mandating the running of the blue running through the Blueprint just to make sure that, you know, um, you augment whatever the human has come up with.

Um, in addition to the human led I and Blueprint, what we work with Pega was we gave them our mainframe source code, which is COBOL. It's not even straight COBOL. It's converted COBOL, which is Idms, like Idms code converted to COBOL. So it's very, very, very confusing. So but we gave that code to, um, Pega team, and then the Pega lighthouse team was able to take it, and then they fed it into AWS Q for developers. And then they worked with AWS, and then it generated a lot of document.

When we looked at the document, it was very impressive because it didn't cover everything. It was not 100%, but it was a good leg up. Before that, what we do currently is we have people in our mainframe team that actually read line by line of the code, and then they produce This document that's then get handed over to the business team so that they can understand that, and then they can write user stories. But now with this AWS Q for developer, we whatever Pega has done, we feel that it's such good. Progress that at least a good maybe 2,030% I cannot put number yet, but a good percentage of work is already done for us.

And then Pega took that document and fed it into Blueprint, and then was able to generate an application for us that was really impressive. And the application was good. But, um, it is it's not yet ready for production, but it was really, really good progress and we are very impressed. We don't want to compare, uh, what we did here with the chess program and the problem that, uh, Alan was trying to demonstrate, you know, versus Stockfish. Uh, but we do see great promise in that technology. Um, and we have to be careful. That's why we need to have human interactions, review some of the code, review the output. And is this really what we're doing right. And once we get to that confidence level, I think you can then move it on to the next stage.

Um, this is the lighthouse we were talking about. So Pega reached out to us and said, hey, we want to do this legacy transformation journey with you. Uh, would you be willing to partner with us and new Jersey quotes? We are always more than happy to partner. We have our CIOs. Uh, blessing. He's always enthusiastic. He wants us to do more. So we've done, um, this, uh, the Legacy Modernization initiative. And then we are also going to be partnering with them for Process Mining. We have very complex workflows. So we will be giving Pega our data. The three fields that we are asking for, and we will be evaluating the Process Mining and see if it's a good fit for us.

We are also extensively using Agile Studio um for our Agile Scrum methodology. We put all backlog user stories, everything there. So we are working with them to, um, get APIs so that can feed into other systems. Um, all I can say is Pega is extremely helpful. The technical team, um, they helped us with a lot of things, including extracting the APIs. And, um, it's been a really good journey.

Um, yeah, it's been critical to have the right people in the room is really what it boils down to, right? So when you have the right people in the room asking the right questions, the right output comes out. And Pega has been, uh, very good with us providing the right people from Cloud. Subject matter experts. Blueprint. Subject matter experts. Workflows. Prpc. Name it. They've provided all of those subject matter experts all in a room together, and we're able to do great things. I think that's about what you guys have right now so far.

So we're going to take some questions. If you guys have questions I think there's I really can't see anything up here. There's mikes on either side. You can come down, say your organization, say your name and ask a question.

I will kick it off. Um, because I think people might be a little shy. Um, but as you guys have been kind of early adopters of a lot of the things that we've worked with together, especially with Blueprint now, you were sort of raising your hand early on to say, let's see what we can use this for with legacy transformation and how we can help modernize these systems with Blueprint, with the things that you see coming out today and some of the technologies that we've talked about at the keynotes this morning, where do you see the value going forward? What areas do you find? Blueprint and I lead LED transformation technologies really making their mark for for what you guys are trying to do.

Right? So there's a couple things. One, just looking at Blueprint and the roadmap there. It has grown tremendously from 18 months, 14 months ago whenever it was first launched it to today. And I'm sure there's going to be more coming down the pike. That will be exciting, right? But also the injection of something like AI and having the Agentic Other Process Fabric. Right. These are some of the things that maybe we can build agents out of Blueprint we can give it. Here's our business. Here's our use case. What agents do you recommend? What agents can you build? So I think that it's it looks promising for sure. Amazing.

I would first analyze the mainframe and make small pieces of it. Identify your. If it's your first time, I would identify less critical business critical piece of it and then identify what is the which technology is the best fit for your for that particular requirement. Is it Pega. Is it AWS. For us we that's why yes, 90% 90 plus applications are in Pega, but not 100% of our applications are in Pega. We go by the technology. Pega has worked for most of our cases, so I would identify first do the analysis, identify what is the best technology, and then slowly move and definitely use Blueprint and the success. You will be able to get good output from Blueprint only if your prompts are really good. So I would spend a lot of time on the prompting trying to go through different iterations and then finally understand what it is that works for you.

It could be a financial decision as well. Part to to your decision making process, right? How much are you paying for your MIPs or your Msus? How much licensing are you paying for certain technologies? Can that come off? So again, it's a case by case question and it's a matter of what's more important.

So we're about out of time. I will quickly plug. There's a lot more out there on transformation. There's a few examples of this tomorrow. Obviously, Don's keynote won't be something you want to miss. Some stories out in the breakout sessions come find us in the Innovation Hub and I can speak for all of you here today. I think when I say that, when I think about new Jersey from now on, it's not going to be Bruce Springsteen. It's going to be Edwin and Lakshmi from new Jersey courts. Thank you so much. Thank you guys for joining us at PegaWorld. Thank you everyone.