Passer directement au contenu principal

We'd prefer it if you saw us at our best.

Pega.com is not optimized for Internet Explorer. For the optimal experience, please use:

Close Deprecation Notice

PegaWorld | 45:35

PegaWorld 2026: AI-led Mainframe Modernization with Pega and AWS: Accelerating Cloud Transformation

Enterprises have spent decades (not to mention millions of dollars) trying to shift workloads off of mainframes and onto the cloud – but the effort has been fraught with obstacles. Enter Pega Blueprint™ and AWS Transform: agentic AI tools that allow IT leaders to reimagine processes as cloud-native applications and retire legacy systems. Now, legacy transformation represents an ideal use case for AI agents. Prompts? Where we're going, we don't need prompts!

All right. Like I said, I'm going to set the stage a little bit. So let's let's dial things back. The year is 1987. Ronald Reagan is in the white House making America great again. Tom Cruise is in the movie theaters starring in Top Gun. And Gary is in a corporate data center. Gary is doing something magical. He is writing COBOL, and he's very good at it. He's documenting business logic that's taken decades to understand claims processing rules, onboarding logic, actuarial claims. So exquisite that the whole business runs on them. And Gary is doing it all in a program, in a system in a mainframe that is built to last. And oh, did it ever last? Now the year is 2025. Donald Trump is in the white House making America great again. Tom Cruise has also starred in a Top Gun movie, and Gary is still in his data center and he is still writing COBOL.

And after all those years, the company is still running on COBOL. But a lot has changed. The code has become complex. There's a spider web of integrations and logic and rules that Gary alone can understand. But pretty soon Gary's going to retire and he does. Some poor fool who wasn't alive when the mainframe was first brought into the business is going to be stuck trying to pick up the pieces and figure out how to run that code, how to build it, how to integrate it. Business is still running on it, but the technology has moved on. We're in an era now, and this organization is stuck in the past. So that's what we're here to unpack today. I'm pleased to be on stage with two folks from AWS who are experts in this world, Ujwal Bukka and Sourav Sarkar, who are deeply involved in AWS transform.

You saw Omri on stage at the keynote. He gave a great rundown of what transform can do. We're going to get deeper into the weeds here and go into exactly how the process works, how we go from that mess of mainframe code and COBOL stuck deep in deep in your enterprises, into an agentic architecture that's built for the modern era. So with that, I will hand it off to Saurabh first to get us started. And I'll click ahead for you and you can take it away. Thank you Sean. All right. So let's start by understanding why do we need to modernize mainframes. So when we talk to our customers about the challenges they face with their mainframe applications, we commonly hear this broad themes. First one is cost. So as you might know, mainframes are expensive. Beat the cost of the infrastructure or maintenance, and even the costs for third party tooling, which is going up exponentially.

Right. But when you basically migrate and modernize to these more Pega, AWS composable solution, you get to choose from a flexible pay as you go pricing, and you can do unified toolchain procurement, which can drastically cut down your runtime costs at the extent of like 60 to 70%. Uh, next one is agility. Like, as you know, mainframe systems are highly monolithic and heavily intertwined. So for you to even make a small future change, you have to go through cycles of testing, which typically runs for many months, right? And that is the challenge here, right? But when you basically adopt to this, uh, modern approach, you, you can basically easily apply your DevOps is ops practices alongside that because it's reimagined. You have full flexibility to adapt to a new architecture, right? Be it separation of concern or microservices, modular design and whatnot, which reduces your blast radius and basically brings in agility to your system.

So you are essentially able to release new features to production in days. Next one is innovation. So there are many capable services available today. But as you know, the challenge is integrating them with an existing mainframe application. Right? So but when you adopt to this modern solution, you can access over 200 plus services available in AWS cloud. And with that, Pega also comes up with a lot of genetic features and solutions that you can easily plug into, right? And with that, you can innovate much faster, much quicker. And finally, is resilience, right? So typically, if you think from mainframe side, this infrastructure's where provision, keeping the peak workload in mind, which often leads to wastage of resources, right? But when you basically modernize, you can adopt to high highly available and elastic architectures through which you can become, I mean, your system becomes resilient by default, right?

And also, one more point I would like to talk is like on the mainframe side, uh, typically the issue resolution is more reactive by nature. When it happens, you try and go fix it out. But when you move here, you, you typically can deploy automated runtime health monitoring. And with that, before even it happens, you can take actions on it. All right. So now let's try to understand, uh, re-imagine pattern, right? So we have about 7 to 8 application modernization patterns available, right? But when we talk specifically about mainframe modernization, that too, from the lens of, uh, DC exit, there are only few patterns which stands out. Uh, typically Replatform refactor and reimagine, right. And reimagine is the most transformative pattern of. All right. On the left hand side, you can see a mainframe application. Uh, business functions are deployed as application code, typically COBOL, right?

You have databases like DB2 and you have application middleware. Maybe MQ is running on mainframe operating system and hardware. Uh, but when you see the right hand side, each and every block is green, which means it's modernized and changed. So business functions, you don't necessarily need to inherit the tech debt, right? You can change and tweak your architecture the way you want and adapt to the modern workflows that Pega offers. Your application code is deployed to Pega industry leading low code. No code platform. Right. And you get a managed runtime managed database, which all is running on Pega platform, which runs on AWS cloud. And integrating DevOps sysops practices with these modernized applications are very easy, as you know. So this slide talks about the reimagine approach end to end. And we kind of segregated it into three distinct phases reverse engineering and discovery, forward engineering using AI and finally deploying it into Pega.

Cloud. Right. So let's double click on each of them and try to understand what happens here. Right. So reverse. So we will take the mainframe use case here. But this can be adopted. This this solution can be adopted to other uh text text as well. Lotus notes, BB and many others are coming in, right? But we'll just try to explain the approach with mainframe use case. So you basically in the reverse engineering phase, this happens in AWS transform for mainframe, wherein you take your legacy application code like COBOL, JCL, DB2 schemas and whatnot. Right? And you basically feed into transform and run, execute a job in transform. And in AWS transform will basically run multiple steps like code analysis, data analysis, business logic, extraction, technical document generation. And each of these agents will basically focus on each and every different different capabilities, right? And all of them will do is go and analyze your code base and will generate some facts, which is essential for you to proceed for the next step, which is the forward engineering, right.

We'll do a deep dive later on. So as you can see at the bottom, the facts are coming out like the integration artifacts, business logic, workflow mapping, and whatnot. This includes both functional and nonfunctional aspects of your application. Then you go to the next step, which is forward engineering. So Pega and AWS jointly have developed agent. We named it as Pega Blueprint agent. Interesting part. It is natively integrated within AWS Transform console. So you have the flexibility to trigger this agent right from AWS Transform Console. After your reverse engineering is done and what it does, it basically enrich, uh, the artifacts that are generated by AWS transform, uh, also synthesizes it and makes the handoff for the next stage, right? Now the next stage is happening in Pega. Blueprint, uh, Pega industry leading Blueprint platform wherein you can intuitively develop your Workflows. You can visualize them even before deploying.

You can see how they will look in different channels such as mobile, such as browser and whatnot. Right? And it's a human in the loop. You can iteratively design your rules, processes and everything. And finally, when you are done with this, you confirm it creates a high fidelity Blueprint, which is essentially the deployable artifacts, right? So and then you go to the final stage, which is deployment. You take this high fidelity Blueprint and go and deploy it in on PegaCloud. And that inherently basically, uh, provisions all the infrastructure for you, right? So your application is ready up and running. And this Pega platform is running on AWS cloud. Okay. So as you have seen, the first phase is reverse engineering, right, which happens in AWS transform. So we will basically double click on this reverse engineering fails to understand what is happening inside each of these steps.

So as I said, the first step is code analysis. So objective here is that you bring your Legacy code base AWS transform will analyze it and will discover many facts, right? First is dependency mapping, right. So AWS transform deterministically figures out the dependency between each and every node within your application. Like program A is calling. Program B, program A is using this database, this file and whatnot. Right. Which becomes your source of truth for the remaining part of the modernization. It does risk mapping. So what it means is like in your code base, let's say you have accumulated as many files as possible, but there are, let's say 100 files which is missing. So transform will figure out and give you a definitive list of elements that is missing. Right? So you go pick them up and basically make sure your context is complete before you proceed to the next stage.

Complexity driven planning. So transform calculate cyclomatic complexity for your programs, which kind of gives you an idea how complex or how easy is this specific, any specific element is, right? So, so so basically that's what principally happening in code analysis. And then we focus on the data aspect of it, right. Similarly. So on the data side we focus on two main things. One is data lineage. Second is data dictionary right. So but by data lineage what we mean is essentially the data flow from where the data got originated, where exactly it is changing and eventually where it is landing the whole data pipeline. So you will capture relationships like, let's say there is a database, right? Which are the programs that I mean, using that data, sorry, which are the programs that is using that database or file in which mode read write right. And from different views, point of views like, uh, one file which is used by which processes one table which used by which processes there is a JCL how many files or databases it is interacting with in what mode.

So all those relationships will be captured, normalized and will be saved as a is a standard artifact as a foundation, right? On the other hand, Data Dictionary focuses on the schema side of things, which means the data structure. So you might have a table in mainframe side, you might have a copy book, right. Or a working storage, which has a database which has a working storage structure, right. So all those structures will be extracted automatically and will be normalized and documented for you. So these two type of data analysis artifacts forms the basic foundation for forward engineering that will use later document generation. So the objective here is basically you have your institutional knowledge embedded within your legacy code base, right. And typically you would go and do it manually, right? So this process is fully automated. You feed in your COBOL programs, your JCL scripts and run this step.

It will create automatic documentation for you. You can control the level of information in this document if you want it for business. You can select a specific option so it will write it in a business language. If you want it to be more technical. Right. You can also do that right? There is a separate option. If you select that it will bring in more technical details right? So basically you institutionalize institutionalize the whole documentation phase. There is another interesting aspect here is like if you opt to do so, this documentation can be embedded, I mean ingested within an integrated knowledge base within AWS transform. And this gives you a superpower to ask any question from the documents that you have generated. Finally, this is the most important feature for emerging pattern. You might have heard us talking about reimagine, reimagine, reimagine. So this is the foundation.

This feature is the foundation of everything business logic, extraction. So we all know this business logic is implemented within this COBOL programs in mainframe specific way. Right. So we have to. So the idea here is we want to differentiate the business intent from the technical implementation, extract the business rules in a way which can be understood by anybody, especially business. Right. So in plain English, we'll capture the business rules from the COBOL programs and other elements. And we'll document it, uh, for you. Right. So this is the main main piece. But apart from business rules, there are many other aspects also which are important, like process flows, dependencies, key functionalities, which will also come out of the same step, right? So this, this with these business rules later on will go and will redesign and reimagine it into Pega workflows. So, so so these are the four basic steps we have used as part of this composable solution.

Now in this architecture diagram, we are trying to show how Pega and AWS have built this Pega Blueprint agent and how it can be implemented and adopted to any customer, right? So let's start from point number one. As you have seen, like we from AWS have provided some primitives to Pega, and we provide it to any partner who is interested, like strands agents, uh, some SDKs and base agents. Right. And then the partner will bring their own expertise and bring their agent spin up their agent on top of it. Right. So Pega has done that. And after that, it is containerized and it is ready. Right. So in the second step, as you can see, let's say a customer is willing to use this product. So so basically we will go and deploy this agent into the agent core runtime environment in the customer AWS account, right?

So everything is running in the customer account. They have full control over it, right? And then, uh, the third step is agent registry. Now as part of this, the agents are running already in customer account, but we go and register in the AT&T agent registry. AT&T is AWS transform. Uh, and once you do that, then AWS transform can discover this Pega Blueprint agent, and it will show up in your console, right? And, uh, as you can see, then the final step is you use, go and use this agent, right? So you go chat with AWS transform, click on this specific option. And it basically runs in the AWS console itself. So that's the end to end piece. And with that I'll hand it over. Yeah. I think, um, if we progress, hopefully this video will play smoothly and we'll see this. This is sort of a zoom out view of the whole thing that we just went through, and a bit of a voiceover for how it looks and how it runs from end to end.

Most mainframe modernization projects fail to take off. Pega and AWS have developed a methodology to change that. AWS transform analyzes mainframe code to generate and enrich technical and business documentation, while Pega Blueprint reimagines processes as cloud native AI ready workflows. Let's see how it works. We'll start with the mainframe application for credit card management. AWS transform can analyze the underlying COBOL code to generate technical documentation and extract business logic. The user can access a dashboard readout summarizing the job. Next, AWS transform can call the embedded Pega Blueprint agent to start reimagining. So just to pause here, um, based on what Sarah has said, you can see that in the previous step with using AWS transform, you were able to extract the business logic and all that good stuff. And then after that, within the AWS console, as I mentioned, since Pega has already built an agent using the composable framework of the AWS transform, you can invoke the Pega Blueprint agent, which you just saw.

And now the Pega Blueprint agent in the back end is doing the job of the rest of the job of the transformation here. Processes and logic as a modern workflow application. Pega Blueprint is an AI powered workflow design service. It can read Legacy system process documentation and generate modern case types, business workflows, data models, and offers the ability to validate business rules. Users can refine data objects, integrations, and personas with Blueprint. After building out the Blueprint, users can preview their application and experience the UI through various channels, including the desktop, mobile, self-service portal, and more. Systems of record can be migrated off the mainframe or Legacy. Data can be accessed from within the Pega application via open Legacy API. The Open Legacy API handles integration decoupling, generating a migration package and reverse API without touching the mainframe. Use the high fidelity Blueprint to deploy to Pega Platform.

Now. Credit card account management is deployed to Pega Platform, and data is managed locally within the Pega system of record. Meanwhile, the credit card domain stays on the mainframe, but is also managed by Pega, which accesses the mainframe system of record via open Legacy reverse API mainframe modernization with Pega and AWS is phased, low risk, and built for speed. Business continuity stays intact while your teams move forward. Not someday, but now. So a quick high level view of that. And thank you all for for adding that context in the midst. And this is also something you may have seen at the mainframe demo on the Innovation Hub. If you've if you were there, um, as they're running a demo of very similar to this, but I'll also just introduce a quick bit of additional tooling and I think maybe another click to. Yeah. So, so we've built this tool in-house here at the Legacy transformation practice team at Pega.

We're calling it foundry. Essentially, this is built to facilitate that process of going from transform into Blueprint to basically give you more control over the kinds of data, the integrations, the business rules and the case types and personas and whatnot that you would bring into Pega Blueprint that you would push into the Pega application. Essentially, we're looking at that output from transform in this video and saying, hey, you found 700 business rules. Maybe you don't need all of them, but let's go by the specific domains that they're located in. You can also trace the business rule back to the AWS transform output where it's located. So it gives you great traceability. We've heard from clients that are using Blueprint to transform, but they want to be able to see. I've used AI to identify business rules, but how do I know where that comes from?

This is a thousand page business requirements document. Well, this makes it a lot easier to do that. And it basically just makes it more deterministic so that you can, um, have control over what you're porting into Pega and what you're leaving behind, or maybe waiting to transform in a subsequent iteration. So that's just an extra bit of tooling that we're using to help facilitate this process. But usual, I think I can hand it off to you and you can keep keep moving here. Sure. So so far, what you have seen is how this particular solution can help you to modernize your workloads and then got in a high level overview of the architecture. And then you have seen a demo of that. So now the obvious question here is that where should I use it? What are the use cases which I have to use this for?

So that's what I'll try to address here in this particular slide. So just, you know, off the bat as a thumb of rule of thumb, if you have any workloads, which is business driven, rules and process flows heavy, you know, this would be a good solution, which you could leverage for modernizing those workloads. And here are the some of the business use cases specific to the particular domain, which you could leverage. And I'll try to double click on each of these things just to give paint a high level picture of how and where you could leverage it to start off with. For the banking and financial services, you can use it for loan origination, product lifecycle management, or account lifecycle management. These are, as we all know, legacy COBOL applications, which are running on mainframe legacy applications, which you could modernize using this particular solution.

If you move to the Insurance domain, some of the examples are the use cases would be policy management claims orchestration, or underwriting. These are rules heavy workflows, which again can leverage this particular solution to modernize them. And then for the communications and telecom domain order management billing dispute management, these are heavy volume process intensive workflows, which you could again leverage them to use the model. So I'm trying to stress upon the key words when I'm trying to wrap it up, that you use case so that it will build the rings when you, you know, when you go back and try to find those use cases for the health care domain. Um, some of the examples would be, you know, a prior authorization member grievance and product entailing here. Um, these workflows are highly regulated workflows. Again, rules driven. So those would be the good use cases which you could leverage to modernize.

And if you move towards the government sector, some of the examples would be, uh, citizen service delivery benefits administration and grants management. Again, the pattern is same here. These are a bunch of complex rules running on legacy old systems, which where you could leverage this particular solution. Last but not the least, um, Manufacturing domain. Some of the use cases would be engineering change orchestration, warranty claims processing, and then uh, supplier onboarding process. Again, if you take a step back here, if you pick any use case, if they have a embedded business logic process orchestration, or business driven workflows. Those would be the right candidates where you could apply this particular solution. And then now I would like to introduce you to a high level paint picture about the modernization platform, which Pega has put together, where you see all the things which we have been talking about, how they all come together and help with your modernization.

This is that platform, and I'll try to break it down layer by layer. At the bottom of the layer. What you have is a discovery, analysis, and pre-processing phase. That's where you would leverage AWS transform, where you point out to your code base, which will extract all the business role and all the goodness out of your particular source code. And then as a part of this layer, you will use the Pega Blueprint agent, which was built by extending the AWS transform extensibility framework, sorry, composable framework, and that that Pega Blueprint agent will take the output of AWS transform and does the enriching on it, and then prepares it for the next level. That is where the Blueprint application comes into play. That's where we're showcasing that in the next layer, that AI, AI powered application design and generation where the Blueprint will take the output of the previous layer and then tries to generate the no code, low code enriched workflows, which you can tie that back to the your business logic or your extracted rule from your code base.

And then these workflows run on top of the Pega platform, which runs obviously, which runs on AWS. And then this Pega Platform has all the bells and whistles through which you could do any meaningful integrations. That's what we're trying to represent in the last top layer in the data and integration here. One, one point I would like to make here is that what we are seeing here is that we do not be doing the Big Bang, whole shebang of your whole mainframe applications, from your existing system to the new system. But what we are saying is you should piecemeal those things and use case by use case. You should do that. And when you're doing that, obviously the other challenge is the data challenge, where some of your data would be in the old system, and some of the data would be in the new system, where that's where you can leverage open Legacy.

That's what we're trying to showcase in the demo. Um, the Pega Platform is flexible enough, or this particular platform is flexible enough where you can bring in the open Legacy, which ties up the data for your existing system, the old legacy system, and also the new system. So the data can live on all, both the places. But again, at the same time, you can case by case, you can bring in the new use cases, modernize them and push it to the new platform and the left hand side. What you're trying to showcase is this whole platform is powered by AWS and the right hand side, uh, again, trying to showcase that this platform is flexible enough where you can bring in your own solutions like open Legacy or your own custom solution or a GSA solution. And you can also leverage their solutions or Sayan Ghosh partners for driving your modernization.

And then, uh, next, uh, I want to introduce you to one of the mechanism which we are leveraging to bring the solution to the customers out there. This is a mechanism is called experience based acceleration. This is an AWS mechanism. So what we did is we worked with Pega team and then try to bring build a joint experience based acceleration mechanism. And what that mechanism basically has is a high level, three level steps where we in the first step, we do the executive alignment in the sense we sit with the executives and try to understand your business goals or organization drivers. Basically, the the intent of this particular step is to get the executive alignment and understand which particular workloads or which particular things you want to modernize, and then you move to the next step called readiness or prerequisite step, where we, we try to work backward from your business goals and try to bring in all your stakeholders and try to understand and peel the onions and do the reverse engineering of which specific use cases you want to modernize and which are the specific things which make sense, and what are the roadblocks of that, and how could we unblock that?

And with the help of the stakeholders, we try to unblock that and then move towards the EBA. Or it's also called as EBA party. It's basically a two day workshop where we bring in all the stakeholders, which we have identified in the previous state, and also bring in the use cases and start to work on that use cases using this particular solution. When I say particular solution, this Platform solution, and then try to modernize that work workflows. So the end of the two day workshop, you will have a modernized application for the identified use case. So now if you take a step back and if you understand what we did here is not only we have a modernized workflow, but you have also defined a mechanism where you are, you are your team knows how to reverse engineer the things, how to leverage this solution.

So now you could use this mechanism to as a rinse and repeat, and to carve out more and more use cases out of your Legacy application and run through the solution and get a modernized application. Um, so again here, one, one call for action I would recommend is that if any, any, any of your team are looking to modernize your workflows, feel free to reach out to our, any of three of us or to the Pega team where, you know, where we could leverage this mechanism to start your modernization journey then. Yeah, yeah. And as you guys may remember from yesterday morning, we had Sheila from Unum on stage, the chief information and digital officer, talking about their journey modernizing their colonial claims application, using the using the methodology that we just outlined today with transform, with AWS, and with Pega Blueprint and into Pega Platform.

And I, you know, obviously that session, if you missed it, then it's absolutely will be, uh, on the PegaWorld website, you'll be able to watch the replays and everything. So definitely go back and check that out. Um, but as a quick high level, um, they, the headline for me on that was a quote from a vendor for seven years, $25 million or something to do like an equivalent or perhaps big bang approach to modernization. And, uh, I don't know if any of you guys remember seven years ago, very different world seven years from now. I don't trust anybody who says that they're going to do something in seven years, because we have no idea what the world will look like by then. So take it iteratively. And that was the approach that we took with with Unum. And there were some preemptive results, predictive results here and thousands of hours, developer hours saved in manual work, um, consolidating seven screens down to one consolidated view, a much more agile and efficient approach to claims processing.

Um, and it proved out that this process really works. And I think that they were really pleased with the speed at which we were able to get a brand new, fully functioning application off of COBOL and into into Pega Platform within three months. So that was an awesome use case. We've got many more on the way. If any of you are looking at modernizing your COBOL application like usual, I'll just said, we'll get you into an EBA and we'll, we'll work with you to, to move across this, this method that we just, we just outlined. And that's about it. I feel a lot better now managing my, my enterprises COBOL application now that we've transformed it. So, um, we have plenty of time here for any questions. I will actually canvas once more. Is anyone from open Legacy here? I thought maybe perhaps 1 or 2 folks.

And if not, that's fine. I just want to make sure because we did mention them in the demo and, um, they're able to handle a lot of integration and data migration. So I'll open it up for questions. But maybe, um, to start it off, I'll just ask either of you to double click on that a little bit. Because at the booth, one of the questions we often get is, okay, you can modernize a workflow, but there's all this data attached to it. There are integrations attached to it. Either of you want to want to just maybe go into, go into a little bit. I can. I can cover that. So basically, as part of your modernization, one of the main question is like, what will happen to my data? Right? So we are so as usual said, the pattern we recommend is of course not big bang, right?

We recommend you to adopt strangler fig approach, right? So when you run your code base through AWS transform, it basically gives you business domains, right? And within each of these domain, you will have different workloads packaged together. And AWS transform will do that piece for you. But the point here is like now each of these domains can be a scope for migration, right? So let's say you get ten domains out of your application. You take domain one, modernize it to Pega and put it to Pega. Right now, you have to take a call during this modernization that how how do I want to deal with my data? Right? So you have two options here. Pretty flexible if you want to take out, uh, maybe some part or full part of data and host it in Pega, you can pretty much do so, right? It scales out on the infrastructure of AWS with services like RDS and whatnot.

But apart from that, there could be use cases also. You might want that. Okay. For time being, let the data be there at mainframe, right. But in order to application to work, you need to build some integration, right? So we wanted to double click on this particular point how that integration will work. Right? So we have a partnership with open Legacy. Uh, so what open Legacy solution does is like they can expose the mainframe services as API, right? And your Pega applications are natively API anyways, right? So you can integrate these APIs, make calls from mainframe to Pega or Pega to mainframe so that your application works seamlessly, right? And over the period of time when you bring your whole database to Pega. Right. Eventually you will dispose these integrations. But for you to complete this modernization journey, right, you would need these integrations, uh, during the pace of the journey.

So that's how the database works, right? Cool. Any other questions? Feel free to raise your hand or come up to the mic. If you have anything specific you'd like to discuss, please feel free. To grab the mic. Yeah. Hi, I'm Kishore from First Horizon Bank. Um, and like in a demo, it seemed very, um, smooth. Um, is that a similar kind of experience or do you think there are these weird Legacy anti-pattern messy code sitting there? Probably that is something we may not be able to deal with. Or are you like 100% sure? Like any COBOL code, however messy it is, we'll be able to modernize it. So I can start off with. So one thing is like, you know, using AWS transform, you could get the, you know, the business rules and the business logic and everything out of that. And the next obvious question, the customer, like similar to what you're saying is like, you know, I have a lot of code.

How do I prove that it has extracted all the business rules for that? Yes, you could leverage AWS transform as some of the features of AWS transform, which you can tie to that too. And to add it to that, as Sean mentioned, Pega has put together a page called a Pega foundry, right? It's the name is Pega Foundry. Yeah. So the Pega foundry, which through that you can double click on that and go back to the which business rule you do that. So the challenge there is like obviously as a customer, you would have to cross check that and cross validate whether it has extracted all the business rules out of that and everything. And once you have this information, when it goes to the Pega Blueprint and when it generates the workflows, you will have a way to tie it up to see that, hey, this is the business rule they have represented in this way.

That way you can build more confidence of it. Having said that, again, obviously there will be corner cases which where you might have to go back and see that it's not always a 100% or as such. But but there will be some corner cases which depending upon your your specific use case, you might have to peel the onions there and try to see how we could accelerate things. But yeah. And I would add to that also, just, you know, we're talking about applications that have been in use and had added to them and built around workarounds for years and years and years. If you have perfect, clean, beautiful, untouched COBOL, then you may not want to modernize it. It might be just fine. So the whole use case here is that you have this mess of code that's been built on for years and years.

And so I think that's, that's sort of exactly the kind of work that we're going after. And, you know, we're doing this in conjunction with the AWS folks, Pega folks, your folks, all the business in the room together, it in the room together. Um, you know, it's a faster process for sure, but it is still a work, a workshop. And we still want to validate everything as we go. So that's where that NBA mechanism, or the thing which brings a structure to the chaos in the sense what we are trying to do there is that rather than looking at all the things, try to narrow it down to a specific use case and run through this process. That's where we will learn that, hey, this is how I can tie it up to the back end business rule. I know this looks good and everything, and then you go and see to the end of it and go to the next use case.

Another excuse might have a different challenge as such, but you have a foundation of how to do that and smooth the things a little bit further. So yeah, I'd like to add one more thing to that. I mean, So as we understand, when mainframe codes are very complicated at times very scattered. Right. So I heard this question from one of my customer who said, like you ran business logic extraction. Now there are like thousands rules right now. I don't know, how do I trace it back right to my legacy code, right. So we heard that very seriously. If you have observed, like our AWS General Manager Omri Kessel keynote yesterday. So we are bringing in a brand new feature called traceability, right, which will be launched next week. Right. So what it means is like, for example, we give you 100 rules, right? And then those rules will be implemented in Pega.

Right. But at any point of time time during the journey, you may have a question where did this rule pop up from? Right. So we are trying to answer that question in a definitive manner. So we bring in traceability that we don't just create rule. We bring in the exact code snippet from where the link from Legacy code, where the rule was extracted from, and we pass on that context to Pega. And Pega has built their own solutions, and even in Blueprint and even in later stages. So every part of the journey will carry on this context, right? So you will have like a mechanism to query wherever you want. If you want to do it in transform, you can check it there. Pega will have an intuitive rule based search, right? You can search it right at the end of the modernization. You can do it.

So every phase of the life cycle is going to basically support that. So that's going to come out pretty soon. Okay. Thank you. I'll take the luxury of asking one more question. No. Sure. Yeah. Um, any banking services like how, how much of banking services already are taking advantage of this? Like, can you give any references? Uh. Can you repeat that? How many banking services are we working with for this? Yeah. I mean. Yeah, we have a few. Banks. That are in the, that we're engaged with actively for, um, mainframe modernization. Um, nothing that's finished or gone live, but there's definitely a few financial services use cases. I think we'd be happy to talk you through what we're doing there. If you're, if you're interested. This is more towards, um, Pega 26 or is it also. No. So it is nothing. It you can leverage that, uh, right now also, um, I got 26.

What would be you have an MCP servers, a few of the things which you could leverage Agentic stuff, but whatever we showed today, you have, you can leverage that right off. Either Blueprint is a brand new, you already have a Blueprint. Blueprint is. Yeah, they've been they launched the Blueprint in 2020 for the Blueprint is not something launched today. Yeah, yeah, yeah. And yeah, so the banking sector also, there are definitely, as Sean said, there are. Huge opportunities in banking. Absolutely. There's mainframe COBOL everywhere. Yeah. Anybody else? Questions? Here we go. Yeah. Yeah. Um, I'm trying to ask questions of my mainframe. Colleagues back home. Might ask what a mainframe developer myself, but, um, I wanted to ask, is assembler PL one also compatible with this transformation? Okay. I can take that. Yeah. Thank you for asking this question. So the question is like whether assembler PL one is supported or not.

So from AWS transform perspective we are trying to scale. And you would see changes coming in the platform every time. Right. So we wanted to target z OS first. So currently COBOL, DB2, VCM, JCL six is supported. Uh we recently added PL one support as well. So the answer is yes right. Assembler is not supported. But we got to hear this feedback from many customers. So it's very high in our FR. So sometime this year you can see assembler is also integrated. Yeah. I have one more question. I did speak to one of my colleagues. Sure. I think they said they have experimented with transform about a year ago. Have there been any advancements since then that I can reassure? I think there were some of the feedback was that they weren't in love with the UI or the. Or the design. I can't remember, but.

Or have there been advancements in the last year that I can come back and try to sell it to them? Okay. I can take that too. So so the so AWS transform, although it's the same product, but it's very different as of today. I've been working with the product since its inception as a specialist. So I have seen the journey, how it has evolved. So last year if you would have experimented with transform. So I was saying, if you remember, like there are many different patterns in mainframe modernizations, right? So let's say if you connect it back together, the product was more aligned with refactor pattern last year, right? But now our front end center focus is reimagined, right? So we want to focus on reimagine. That's our vision, right? Of course, we want to support other patterns as well. So right now the product is more evolved, right?

It's more focused towards reimagine. So I can guarantee you if you try it out again now, you will have a whole different experience, right? And the workflow of reimagine would be different than refactor. So you might have to pick and choose the different steps. We have worked through like code analysis, data analysis, technical document generation, and especially business logic extraction. These are the core steps for imagine. So yeah, we highly encourage you to go and try out right to see the differences. And if still there is a, you know, something which you doesn't meet your goal or something like that, feel free to provide that feedback back through your AWS account team. Um, you know, the service teams back there would love to hear that feedback and make it better for sure. Thank you. When was transformed first rolled out. It was rolled out in in May 2025 was the GA, but it was in public preview for like maybe six months.

January 2025 onwards. Yeah, sure. Right. So yeah, I'm constantly changing. Right? Of course. Yeah. Great. Thank you. Anybody else? No. Um, please give our presenters a hand here from AWS. Thanks for joining us on stage here. Thanks for the help. And thanks for coming. I know it's the end of the the end of PegaWorld. Hope you guys enjoy the chess tournament and some cocktails down in the Innovation Hub and have fun in Las Vegas for the rest of your time. And if you have any other specific questions, come find us. We'll be happy to answer. Thank you. Thank you. Thank you.

Ressource associée

Produit

Une révolution dans la conception d’applications

Optimisez et accélérez la conception de workflows, grâce à la puissance de Pega Blueprint®. Présentez votre vision pour générer votre workflow instantanément.

Vous souhaitez créer un Blueprint ?
Choisissez le moteur adapté à vos besoins.
Pour la conception de workflows et applications

Repensez vos processus et transformez n'importe quel workflow en application prête à être développée en toute confiance.

Pega Blueprint®
Pour la conception de stratégies marketing et d'expérience client

Visualisez et activez les parcours client et les stratégies d'engagement sur tous vos points de contact.

Pega Customer Engagement Blueprint™