2447 - Advancing Software Development Through AI and Quality Assurance with Redwerk's Konstantin Klyagin

0:03Welcome to the Thoughtful Entrepreneur. If you're committed to growing your business by serving audiences with generosity and real expertise, you're in the right place. This is a daily 15-minute show featuring leaders who believe real impact comes from generosity, not hype, which means as a listener, you're going to get the good stuff. In fact, we created podverified.com to give thoughtful leaders like you free tools to show up better on podcast stages like this one. That includes your free podverified score, free podverified guest training, a free media kit builder, and so much more. It's the first ecosystem of its kind built by podcasters who genuinely want to help guests thrive and grow in podcasting.

0:46Visit podverified.com if you'd like to be featured as a future guest. I'm Josh Elledge. Let's get into it.

1:20do for fun when they are not at work. And we were just talking about your hobby. Tell us what you've been doing lately at the beach. Well, lately I've been practicing jumps while kite surfing. And this is the next level for me to master this sports. And so, you know, I've had a, I had a friend who was into it. Tell us for someone who has not done kite surfing, um generally like how it works is it is it hard to get going with the basic stuff um is it i don't know just for someone who's never done it give me give me an overview of like why you like it and and what someone might want to know before they start uh start this hobby it depends on the kind of a kind of a person you are so if you're a computer geek like me then it's probably a bit to start riding because i've been trying to master this since 2011 and 2022 i really felt confident riding but before that i practiced that every winter in like in many parts of the globe like i i took lessons in the dominican republic in australia in egypt in florida by the too and uh in colombia if i didn't mention that and uh then the time passes like in a year you forget you forget all these skills and then you have to learn uh all over again but once i learned to portugal uh i did twice a week uh my lessons then i bought all the gear and now i'm i'm like an advanced rider but the next level is the jumps and that's a whole different level next level of difficulty there well kite surfing the world doesn't sound like a bad gig so well very cool well let's talk a little bit about red work uh so you have been uh you founded the company is 21 years in operation.

3:29You've got over 90 tenured team members. You guys are not small potatoes. You've done some really great work, including a lot of government apps. And do you mind just giving us an overview of RedWork and the work you typically do? Of course. So we work either with the tech companies, helping them offload their work, which they find less of a priority than developing new products, let's say maintaining existing products or building modules. Or we work with non-techies and companies that are about to undergo digital transformation, but they don't want to know anything about the tech side of things because they want to concentrate on their main business.

4:16And then they come to us as software experts, and we build their entire technology for them. Wow. Okay. Well, give us, Konsti, if you could, I'm curious how your world has changed or how the development world has changed from your perception, you know, with the popularity of a lot of AI-assisted coding tools. Give us an overview of your world, if you could, or kind of like a state of the union for where we are with development right now. Of course, of course. And by the end of the video, I'm going to share the three questions that you should ask yourself about your software product in the AI era.

4:56Ooh, good. All right, don't forget. Let's not forget that. Of course. So let me tell you about what happens in our business once the AI got advanced enough to turn ideas into actual code. So people started shipping faster, right? So whatever idea you have, You can Vibe code an MVP and even at some extent go live with it. But what I'm seeing and what we are seeing a lot is that without proper software knowledge, without proper architecture, you cannot really ship a full scale solution that will withstand the load, which will work as expected for the customers and that will not break either during onboarding or its main functionality.

5:54So let's say for Vibe coding and for fast shipping of simple things, relatively simple things, the AI is good. But then the more people ship, the more they need software engineers to actually build it for a larger scale. And they also need our second brand is QA Work. So I run actually two agencies, not just one. Red Work is one of them. And QA Work is the other. So QA Work tests the software. And the more the software is shipped now, the speed is getting crazy, like really through the roof. and the more human quality assurance is required because you can engage the AI as well on testing this stuff, but then will it work the same way for humans who are the main target audience for your software?

6:50So people come to us also to have an external perspective on their applications traditionally. And recently, they're coming to us to test their AI stuff and also to redo their AI-generated, vibe-coded stuff the proper way that can be deployed at scale. Yeah, so the availability that regular folks have to vibe code and MVP, do you see that as a good thing for companies like Redwork or, you know, true developers? or do you see that as competition? I don't see that as a competition. We see it as a useful tool to employ in our everyday work. So whenever we need to speed up or whenever we need to communicate with the customer what's possible and what's not, or the customer wants to actually show what they want in the real production product, so they vibe code it on their end and they send it over to us and they show it to us so that we do it properly.

7:57because like uh let me tell you an example from our like experience from a few days ago we were looking at the vibe coded uh code and uh like one of the lines well actually we reduced that uh particular module from i don't know a thousand uh lines of code to maybe if you if you like dozens lines of code. And one of the lines said, declared the constant that equaled max between zero and another constant. So this is like a very typical AI slope, which we needed to clean up. And this happens a lot. I mean, which is fine in the MVP, which is fine to show to your investors, to your partners, to your team.

8:49But then would you really like to have that in production, actually facing your users? Yeah, and, you know, I got to say that, you know, I've seen, I think we've all, I'm sure you've seen a lot of slop. And, you know, again, I don't know if you, you know, again, if it's MVP, great, you know, now we get a proof of concept out there. We could kind of see it, feel it, you know, get our hands dirty with it, you know, realize what we don't like or like. And, you know, however, you know, one concern I think, you know, folks might have is, you know, if you're developing or if you're going to hire a developer, you know, how do you have that conversation?

9:33I think you had mentioned, you know, the kind of the three things to look for, you know, because I want the code to be solid. I want it to be tested. I want it to have gone through the QA. And I don't know if there are many developers out there, especially if you just hire a solo developer. They might just kind of vibe code the whole thing. What are the risks there? The risks that it's going to work as an MVP, that it's going to show some functionality. But then as you deploy it live and, like, let's say, a thousand of users simultaneously will try to use it, it's going to break. It's not going to withstand the load or it's going to have some business logic problems that will prevent those concurrent multiple users to work with the software.

10:28All right. So there can be and will be architectural decisions made without you by the AI. And there's going to also be infrastructure decisions that you didn't influence. You just take it and ship it. And then you kind of see if it works or not. And it's a kind of a gamble. And tell me about some of the projects that you've worked on. I mean, you've got part of the Webby Awards, you're part of Champions of Change from the White House during the Obama administration. You've gotten a lot of or you've helped your clients win a ton of awards and as well as some decent industry recognition. Where does Redwork fit in in terms of like, why would someone want to hire Redwork maybe versus another development agency?

11:26Well, because of our expertise, because of our combined experience of hundreds of years, and because of my own experience with the industry, I've been coding since I was eight years old. Wow. It makes 37 years now. So it's pretty solid and I think in a way unbeatable. All right. So for our friend that's listening, I think you – and Konstantin, you had mentioned something earlier. Did you cover that in the way that you want? Remember when I was like, yeah, let's definitely talk about that. Did we cover that fully? You had three items that you said, make sure. Well, the three items, I thought it was the open loop to finish the conversation with.

12:13Oh, well done. Let's go there then. Yeah, yeah. Just remind me about them in the end of the conversation, and I'm going to reveal them because we need to keep the suspense. Oh, wait, we'll go for it right now. You would like me to share the three things right now? Yes, please do. All right. So if you're having a software product, you should ask yourself three questions to make sure if you're doing it right from the quality perspective. So first, in the AI era, you have to really know if you are doing any of architecture decisions or you just ship whatever AI ships for you. The second is if you have someone test the features that you have in the software internally or externally, or your end users are the first line of QA.

13:06That is important. And the third item is, let me remember that. Hold on. And I don't know if you've got this. I know you have a really great blog here, as well as a number of other resources. So I suspect that you may have some of this there as well. No, no, no, no. I prepared it for this podcast. Go for it. Go for it. I got you. Sorry, sorry, sorry. It's just me. Yeah, yeah. And the third item is if you know where your users drop off and why. Oh, yeah. And starting from that, from that knowledge, you can really start building quality software that lasts. Sorry, I don't do podcasts very often.

13:57So I'm a bit nervous. Ah, that's all right. Your earnestness is endearing. So, Konstantin, so I have your website pulled up. I was just mentioning redwork.com. We have a lot of mutual connections. You have a lot of followers, you know, for a software agency. And so, obviously, you share a lot of thought leadership. And I know, like I said, you've got some decent resources here on your website. to our friend that's been listening to our conversation what would you recommend they do next

14:34so do you mean about our podcast listeners right now that have been listening to our conversation how do they engage with you I don't know if you have any resources for someone that wants to learn more about the work that you do through Redwood where to go for kite surfing lessons of course so the next thing that you can do you can reach out to us about your software for more software development expertise. If you have a digital transformation project going, or if you would like to improve the quality of your software, you can also do that for free by submitting your software product via the form on the website on qaverc.com.

15:19And we will do a back roll for free and come back. We're going to publish it on the website and we will come back with the list of bugs that we found in your software product. Yeah, and that website is... And if we don't find any bugs, you will get the no bugs found page, which would mean that our professional QA engineers examined your product and couldn't find any major issues with it. And that defines your team as a super professional. Your developers are the best. And it's very good to pitch in front of your investors because it means that you are doing your software right. Yeah, and by the way, the website that you mentioned is qawork.com, which is, I would imagine, particularly valuable for those that, you know, again, may have had some AI-assisted coding involved, but that's at QAwork, W-E-R-K.com.

16:28Well, Konstantin Klagen, Slava, Ukraine, thank you so much for joining us. And again, we've got both of your websites listed here. Congratulations on all your many decades of success and again, all of your clients' success and recognition that you've helped co-create with them. It's been a great conversation, Constantine. Thank you for joining us. Yeah, thank you. Thank you. Have a great day.

16:58Thanks for listening to The Thoughtful Entrepreneur. If your goal is to grow your business, increase revenue, and build authority without gimmicks, this show is designed for you. Each episode gives you practical insight from leaders who have turned trust and credibility into real business results. If you value short, thoughtful conversations that respect your time, make sure you're subscribed. We publish daily 15-minute episodes focused on growth that actually compounds. And if you're interested in being a guest or you want to turn podcast appearances into measurable business growth, visit podverified.com.

17:34You can start for free with your pod verified score, along with free podcast guest training and a free media kit builder. These tools are built to help you get booked on better shows, deliver stronger interviews and convert audience trust into revenue. No more desperate sales. You truly can enjoy all the business you like by generously sharing your wisdom. Everything we offer is built by podcasters for guests with a focus on trust, organic growth, and long-term business success. Thanks again for spending your time with us. I'll see you next time.