TLH Ep. 38 This is The New Age of Testing! We talk Automation & AI with Nikao’s QA Director Prav Gudipalli
Hello everyone, and welcome to The Legal Helm! We’re excited to have you with us as we explore the latest trends, challenges, and innovations in the legal tech space. In today’s episode, we’re joined by Prav Gudipalli, Director of Quality Assurance at Nikao. Nikao has been a leader in implementing cutting-edge wealth platforms for major financial institutions, migrating billions in assets and hundreds of thousands of client accounts with minimal disruption.
Prav brings over a decade of expertise in financial technology, specializing in platform quality engineering and wealth management systems. He has successfully led enterprise-scale migrations and built testing frameworks that enhance operational efficiency and drive revenue growth. In this episode, we’ll discuss how automation and AI are revolutionizing quality assurance in financial services, tackle key challenges, and look ahead to the future of testing. Stay tuned for an insightful conversation on the evolving world of legal tech!
Your Host
Bim Dave is Helm360’s Executive Vice President. With 20+ years in the legal industry, his keen understanding of how law firms and lawyers use technology has propelled Helm360 to the industry’s forefront. A technical expert with a penchant for developing solutions that improve business systems and user experience, Bim has a knack for bringing high quality IT architects and developers together to create innovative, useable solutions to the legal arena.
Our Guest
Praveen Gudipalli is an ISTQB® Certified Test Manager with over a decade of experience in platform quality engineering and wealth management systems within the financial technology sector. He has led enterprise-scale migrations and developed robust testing frameworks for investment platforms, incorporating cutting-edge digital interfaces and white-label solutions that boosted operational efficiency and generated significant revenue across the wealth management industry. Praveen consistently redefines quality standards through strategic automation frameworks, leveraging tools like Playwright, Cypress, and AI-powered technologies.
Transcript
Bim Dave: Hello everyone and welcome back to the Legal Helm, where we dive into the latest trends, challenges, and innovations in the legal tech space.
Today I’m thrilled to be joined by Prav Gudipalli director of Quality Assurance at Nikao, a company that has implemented cutting edge wealth platforms for major financial institutions, successfully migrating billions in assets and hundreds of thousands of client accounts with minimal disruption. Prav is an I-S-T-Q-B certified test manager with over a decade of experience in financial technology specializing in platform quality engineering and wealth management systems.
He has led enterprise scale migrations and built testing frameworks for investment platforms, enhancing operational efficiency and driving revenue growth. In today’s episode, we will explore how automation and AI are transforming quality assurance in financial services, tackling key challenges, and looking to the future of testing.
So let’s get into it. Prav, welcome to the Legal Helm.
Prav: Nice to see you.
Thank you.
Thanks for having me.
Bim Dave: Yeah, great to have you here. so, so I thought we’d just start with, um, a little introduction. So maybe you could tell, tell our audience a little bit about yourself, your background, and the journey that led you to your current role at Nikao. Alright.
Prav: Yeah. So, I would start with how I end up here, here being in Nikao. I’ve done my bachelor’s, uh, in computers. Then I’ve done my masters in, uh, computers, again, applied informatics, which has got advanced programming languages as, as one of its core usability methods. And you can ask me a question there, why, why did you end up in testing when you’ve done, uh, lots of programming courses in your patches and masters?
I’ve got , 15 years of experience, more than 15 years of experience, and all this experience is in, uh, financial service organization because, I’ve myself, you know, in, I’ve course lots of investments across the world.
And then I got a lot, lots of interest in playing with the money as well.
When I was working as junior developers, I, I worked closely with this, uh, with these testers at the same time, in the same team. I’ve realized that these testers coming from the business and trying to learn the testing techniques they can only see the UI side of things and they can find defects. But the, the kind of lacking lagging, uh, what’s going to happen in the backend.
I find it very fascinating because when they create a defect, I can, I can clearly see the logic behind it. And I used to do my own analysis, and I used to, because I’m a junior developer, I don’t fix the code straight away. I take it through the senior developers and uh, and, and give them lots of information how to fix this, uh, this bug that’s. That sort of thing kind of fascinated me. And then I’ve got a role as a, uh, as a test analyst on the same team because I just want to progress in my career as well. Luckily the projects were not kicked off yet, and I’ve got somewhere around three months of time too, to go through some lots of internal training that they were giving to all the employees that were joining there. So. This is where I was being trained in a financial services organization, assuming that I was a kind of a BAU person. And I’ve joined with the BAU guys and I’ve, I’ve acquired lots of knowledge in these trainings that, that, that is my starting point in financial services.
So I’ve got lots of knowledge, uh, you know, uh, in, in early stage of my career. And then I’ve, done my test analyst, analyst position. Then later on I’ve, I became a test, test lead and then test manager, senior test manager. Now, luckily I’ve got, uh, an opportunity in an account to be, to become a QA director.
Bim Dave: Fantastic. Well, it sounds, sounds like you’ve been on a, on, on a very interesting journey and I like the fact that you’ve got that kind of mix of the, the core development aspect of it and what can happen at that point, and then how you translate that into. Quality delivery right at, at the end of it. So that’s really, really
Prav: Correct. Yeah.
Bim Dave: So, so in in the, in the financial services space, you’ve obviously worked quite extensively on, um, platform engineering, quality engineering, and the, these wealth management systems. What would you say are some of the big QA challenges that financial institutions face today.
Prav: So if you’re talking about financial services, that which I’ve got experience in wealth management sector, this is where, will, we were offering, uh, different investment products to the clients, you know, as part of our investment platform. SEP, where you actually, uh, store your pension part and start investing it from there. Also, uh, ISRs, GAAS, like general investment accounts, uh, this financial service, products are FCA regulated that’s, that, that makes the financial service complicated because we are always dealing with the client client’s monies and, uh, uh, there are lots of compliance and regulations. Uh, and you know, at the same time, we have to prove that, you know, whatever the platforms that we build. It’s got high performance and se scalability and, uh, o obviously security is the key and also, we need to make sure that we look after, uh, claims data. You know, as per the GDPR compliance, there are lots of challenges that financial service testing, um, faces in this way where these are non-negotiable. we have uh, make sure our platforms that we deliver are performing, uh, you know, according to the FCA standards.
Bim Dave: Yeah, this, I think there’s quite a few, um, parallels actually between, um, the financial services space and legal space actually, because in a very similar fashion there’s a lot of focus on. Client accounts and, and managing client account information and a lot of the data security pieces, which I think are at the forefront of everyone’s, uh, mind at the moment.
Um, so it’s interesting to hear the kind of parallels between the, the financial space and, and, and the legal space.
Prav: Both are kind of customer centric platforms, it is a customer centric business. Like, you know, at the end of the day we need to look after customers and the data and the transactions and we need to provide them with a kind of a, a smooth journey when, when they were interacting to one of our applications. We don’t have space for rounding errors, right? And we have to, uh, calculate these investments them into eight dismal places these are, these are the scenarios that we, we go through daily we need to make sure we understand the product, we understand the regulations of that product. we, as A qa, we need to make we’ve done enough research, uh, on the product that we were testing on.
Bim Dave: it makes total sense. Um, so, so just to kind of dive into some of the, um, the, the kind of testing processes and some of the, some of the technology pieces, um, that are involved. So if you look at the way software testing has kind of changed over the years, um, obviously there’s a big focus on automation and how things have transitioned from being, you know, lots of people testing, um, various test scripts and, and being people driven to being automation driven.
How have you seen that evolve in the financial services space, and what do you think the biggest impacts have been from, from automation?
Prav: So. you know, In terms of financial services we have to generate lots of scenarios and lots of transactions because that’s, that’s what happened in production. So. For me with the team of 10, the time, you know, it’s really hard to create that volume of scenarios. So that’s where we actually implemented automation.
This is back in 2010, 11, where we were talking about, you know, uh, Java with Selenium. And then we’ve created automation framework mainly to reduce the risk of volume. So, We’ve created this automation to create lots of data that that can test most of this, you know, volumized scenarios.
And where my, um, manual team just focuses on the, the business logic implementation, and the key scenarios where we, we identify risk of defects.
Our previous projects I want to mention about as well, we, where we actually followed, uh, BDD approach. We, we only have, two weeks, to implement this project, this change into production. So in two weeks we have to perform all the, you know, system regression and then into production. \ This is where we got the automation, you know to take the risk based approach instead of, uh, conducting this full two weeks, we used automation. We conducted one week full automation, regression. Make sure that starting, we, we’ve, we triggered our regression switch. We, we’ve got confidence on our on, on our production code. We in fact, uh, completed that in, I would say eight days. Uh, rather than like, you know, 10 days is our plan. this is where I’m kind of showcasing. How automation can help us.
And that, that, that we successfully implemented that project and then later on I’ve got that budget, uh, to increase automation into Cas CD pipelines.
Bim Dave: What, what about the role of the role of data? I think you, you kind of mentioned some of the things to consider are the fact that we’re dealing with client data, right. And some of it’s obviously highly sensitive data as well.
Prav: With financial services, you know, data security is key. We are compliance to GDPR regulations so what we do is we anonymize all the client’s information. So, um, as part of testing, what we’ve done is, uh, um, we’ve got some scripts ready where we have to go and anonymize client name, address, bank details, uh, you know, um, contact information.
So, these scripts will be run overnight, and then we, we anonymize that data. Recent days we were, we were actually taking the help of ai, what we’ve done is like recent times, we’ve used the fakers, you know, the, the pandemic fakers, uh, to actually, uh, anonymize um, entire platform with the sensible data in, you know, which is also fake data. that’s one part, uh, of, anonymizing the client’s information across the system. Um, the other part is we were also generating the new and whatever the data that we input to the automation suite. Um, is also an, is also a kind of faker data, but it has called Business Logic implemented in it. So we are using AI to automate all this process of data creation and then inputting it into uh, into our automation suite. So that fee I’ve seen massive you know, benefits in uh, data creation through AI. We haven’t used any premium applications out there. Uh, but then later on I found out there are premium AI powered tools like, uh, you know, functionalized, uh, and then Test Sigma. They’re very helpful, but they’re also expensive.
What we have done is we’ve used playwright. We’ve developed some really good AI based frameworks where it takes only minimum efforts in terms of coding. Uh, and also once we deploy this into production this AI system is learning the way we are conducting the test cycle.
It’s more like AI is giving me information where I can customize the testing suite where yes, you run these as first priority and then all the reusable test cases, all the, all the regular test cases run them as a second priority and then it just goes, runs and provides the reports then and there.
Bim Dave: Yeah. No, this, I mean, it’s, it’s an exciting time, right? With all of the, um, AI technology being applied to various technology use cases, but this, the one you, you’ve just given is a really, really good one. Uh, are there other use cases, um, that you’re seeing or you see in the future in terms of how AI will improve any of the QA processes?
Prav: Yeah. Uh, sure. Bim as I was mentioning you, especially these financial service. , it’s got lots of, uh, regulations. As an user, you have to have, uh, that knowledge and capability of, uh, handling these products we’ve got, uh, different roles like advisors, clients, fund managers, uh, where advisors and, uh, fund managers. They get, they get commission paid on the advisors that were being given to the clients. And also, we’ve got that responsibility where we need to declare what are the future charges that you’ll be paying on this investment. So claims. Demanding lots of information these days. Uh, before it was like, you know, we just generated, uh, you know, a terms and conditions document or a declaration document where it lists all the charges and everything. But it’s not like the clients are now following up because they’ve got AI tools in their, in their hand.
They clearly understand, uh, you know, what would be their growth rate if the growth rate is the point or person’s 0.3 person or four person. So it’s important to make use of all these AI services out there. There are some really good LLMs where we can actually host them in-house, isolate to our Azure cloud or our cloud, and then, uh, make use of that LLM services and then. Calculate all these charges to the point, to the fractional points every time when the report gets generated to the client.
Some other sector where they’ve got lots of priority on the user interface and then the, uh, and the graphics that we’re displaying on the screens, they’re definitely ai tools out there. But now I’m talking about the accuracy of, uh, information, which has got lots of business logic in it, in terms of the products, in terms of their hierarchy of, you know, of company, how it is operating.
Bim Dave: What about some of the, the challenges that AI presents? Right? Because I think, um. While we can see like there’s so many benefits, so many positive things that we can get from leveraging ai.
Prav: So, you know, uh, as soon as this, uh, AI buzzword started, you know, there are so many companies, like we not just open ai, Google, Microsoft, recently the Deepsea the people coming up with lots of different AI versions. At the end of the day, you know, as humans quality analyst we are the drivers. You need to be in a position to understand what was the information that’s being provided by AI’s and how it can be useful and how, how we can incorporate into our framework.
So, you are the owner of the a test automation framework or an application, so. You should be deciding which AI to use or which information to be used into your systems. If you don’t have that knowledge and grip on your framework, you know, you’ll be confused and you’ll, you’ll be lost.
Bim Dave: One of the things that, that we see as quite a common challenge in, in the kind of realms of testing is that you have a firm that will typically take an approach of basically leveraging different best of breed business systems, right?
So they may have a finance system provided by. One party, and then that needs to be connected to a case management system, to a HR system and various other downstream and upstream systems that, that, that all share commonality.
How you make sure that tests are relevant from start to end of, of a process flow, right?
Prav: See, that’s, that’s how I work since the beginning as well as part of financial service systems, we integrate. With lots of other parties to do, you know, due diligence checks, a ML checks, bank accounts checks and all that stuff.
The way that I approach is we have to stick to end to end. You know, this is, this is like you, you need to ensure, like, whatever the journey, you can’t just go and say, Yeah, my client is onboarded, and then that’s it. My, my journey is finished. no, it’s not finished. It’s not like that.
How the client is onboarded, as is the results being recorded properly. What are the transactions that had come through? Are you able to complete the transactions because you know when the client is onboard when we talk about financial services, we do do the due diligence checks like AML Check, bank account check. Whether have you got the results, has it been vetted? These are the small details we have to incorporate on every scenario that we were developing. Uh, and then we need to ensure yes the integration is happened on onboarding correctly. We have to concentrate that as much as possible because if you don’t get the response from the API that you’re requesting, that will cost millions.
Bim Dave: Yeah, no, to totally agree. And, and you, you touched on, um, another important point, which is the, the concept of API testing versus, um, UI testing, right. And UI automation. What, what’s the right balance there?
Prav: So, it, it depends on which phase of testing that we were in. Like, you know, in terms of system testing, if, if I was a, if I’m a test manager and if I’m leading that project, I would I would test the APIs, uh, on a standalone I’ll use the automation we have got in place where we will just give it, give the XML request format to my automation tool. It just, it just takes it and it gives you loads of requests. we just input those requests to whatever, like a postman or so PI that you’re using and then we stress test it in the system testing phase.
When it comes to UAT testing or further down the system integration testing, that’s where we, we have to combine both UI and the API testing because that’s where the integration happens, right?
So that’s where the task cases has to be end-to-end. Obviously people who will be working on these two areas will have that knowledge of how to test by a UI, how to test by API, so they, they know what is the business behind it. Then when it comes to the integration, that’s where we make sure that. Everybody’s spread across all the scenarios end to end.
It’s their responsibility that, you know, uh, whenever we send the request to api, you’re getting a response, valid response, then more to the next step. That’s how we will integrate to be fair.
Bim Dave: And, and so with, with the way, um, a lot of businesses are evolving now in terms of moving away from having a lot of physical on-prem infrastructure and moving into hybrid or cloud, um, solutions or SaaS delivered solutions, does that add a level of complexity in terms of some of the considerations around testing.
Prav: If you’re talking about testing point of view that’s a kind of a good situation for us. Like right. We want to find defects when there is an integration break, you know, we would just you know, being a test manager, what I would suggest to do is like, we learn it from the experience.
I know my automation framework is telling me this is the key risk scenario where the SaaS providers are different in different locations. You just go and test this one first and I’ll say, Yeah.
Just go and test it for tonight. And then it goes and test and, you know, there, there could be some results, you know, which were coming as like not responding in terms of a p and I will get the result and I will escalate it as quickly as possible.
So. You know, these integrations happen, like the, the errors in integration happens, you know, when you talk about development point and implementation point of view. But according to the QA team, we, we need to be there first to identify the issue as early as possible and to address it. As a test manager, it’s my responsibility that everybody understands what was the root cause and how can it be fixed? And before triggering any new releases, this is the notes I’m giving to the, you know, implementation team to make sure that all these different SaaS providers are responding when you’re deploying things. So that’s how we have to mitigate these issues.
Bim Dave: So, um, so with, with all of the, the kind of tooling and maybe some investment that’s needed in terms of the AI solutions that you could either get off the shelf or build in-house from a framework perspective. How, how do you then tie that to measures of success, right?
So you’ve got a strategy to go and implement and go and maybe buy a solution or go and build a solution in-house to be able to, um, automate some of your testing. Are there any, um, key like metrics or KPIs that you typically, uh, recommend focusing on to, to ensure that you’re getting value from the investment that’s being made?
Prav: If you ask me, I wouldn’t just go and recommend any product out there without looking at the system first. We need to. Look at the system, identify the, you know, pain points or inefficiencies and then, uh, gradually we, we just going to introduce ai into the system. Start with small AI enhancements where, uh, and I was talking about self-healing test scripts, right? Like, you know, uh, some of you know, we just need to understand how these automation test cases are failing. You know, they could be failing because, uh, you know, some of the UI element is more.
And then the automation test is failing. That is going to cause lots of headache for the testers. Know, repeating the test again, identifying it slightly mood, and then conveying that to the, you know, developer where the developer or analyst might reject it. Just say, ’cause it’s, it is just slightly moved. So we need to implement self-healing techniques.
I would demonstrate how AI is adding in terms of time, time and efficiencies, and then. Slowly. At the same time, if we are dealing with the team, we need to upgrade the skills of the team as well to, to incorporate ai these guys need to be upskill or trained to understand what they were getting it from our, from the a p you know, ai, and then incorporate it into the framework. So there they’ll be in full control, end to end whenever they’re, you know, uh, test automation fails.
Bim Dave: And, and if, if you, if you look ahead, I think you touched on some of the, the kind of AI things that either in, in progress now or coming down the pipe. Um, how, but how do you see things evolving over, let’s say, the next five years from a QA in financial services perspective and other things that, that firms, um, and teams should be thinking about in terms of how they prepare for that?
Prav: So. Today in test automation is, uh, low code, no code a automation platforms but there are some, some firms out there which, uh, cannot afford, the high premiums, uh, that were being. charged by this, uh, no-code, low-code platforms, you know, uh, at this, at the same time, you know, they want to use AI there are lots of open source tools available out there. Like Apli Tools, helium I was trying to. Build this framework using playwright, you know, and, and all these, you know, open source AI tools. Um, and also having them in-house because at the same time we have to protect the data, data, data security. Um, we have to keep in mind about the GDPR complaints. We can’t just leak the data.
Not all the financial services clients might opt for this premium technology out there. Uh, but instead they might be interested into, uh, you know, playwright, AI automation, you know where there is a, there is a massive community who will be helping, uh, in terms of, uh, any challenges that were being faced in ai, uh, where we can just keep continuously learning about these advancements and then. You know, adapt them into the, the next five years I would say, we will be using AI to give confidence to the client and develop some business applications.
This, this, production log might, you know, the backlogs, uh, our production log issues might, might just reduce, uh, and then the, the, the efficiency, the, the speed of the releases into production will increase.
Bim Dave: It’s, it’s an exciting look into the future pro exciting look into the future.
Yeah. Yeah. So, so I’m just gonna move to some, um, some fun wrap up questions if I may.
Um, first of which being, um, if you could borrow Doctor whose time machine and go back to yourself at 18 years old, what advice would you give yourself?
Prav: When I’m, When I’m 18 years old I was purely academics. I Was into, you know, how do I get job? How do I get, uh, you know, uh, nice opportunity or role development and all that stuff.
But I’ve never looked into future I would suggest, you know, some of viewers, you who are young, uh, to look into the future as well. You know, uh, there are lots of, uh, documentaries out there for example, we were talking about electric cars today, you know, and then there, there’s another stream that was going on.
Hydrogen cars, you know, there, there, this hydrogen is coming from the green source, you know? Some other technologies out there, which were in the early stage today, but they might, they might evolve in future as a, as a, as a game player or game changer.
So, I would, I would also take your pause, look what’s happening out there not just the academics, not just the, uh, the role that you are in. And then, uh, try to understand and, and, and try to spend some time on it, you know, like half an hour a week, you know, just to see, see what’s going on with it, what is the progress about it, and, uh, you know, is there anything that you can learn from it?
Uh, and just keep updating yourself .
Bim Dave: that’s very good advice. Very good advice indeed. And then the final question. What’s your, what’s your, the, the favorite holiday destination that you’ve been to that you would recommend people go visit?
I
Prav: Um. I Would say Rome I’ve seen so many movies obviously series, you know, about, about Roman history and all that stuff. I’ve got some picture how Rome could be, but when I’ve actually visited Rome, I was so surprised, you know, uh, you know, going through those, uh, Roman streets every, every. Every inch of Rome, uh, is more like a, there is a cathedral, like there’s so many cathedrals in one street itself. Uh, so I was, I was very surprised, uh, you know, that, uh, it has, that, it has called Rich Heritage, you know, and Hi, hi History in Rome. So I.would, you know, that’s my favorite holiday destination. I would, I would go back there again and again.
Bim Dave: Excellent. No good. Good recommendation. Good recommendation. Well, Praveen that brings us to end of our podcast. I wanna thank you very much for joining us today. It was very insightful, um, really good tips shared for anyone interested in the testing space. So, I really appreciate you spending time with me today.
Prav: Thanks for having me, Bim.