AI-Powered Tools in Software Testing: A Deep-Dive using Katalon with Mush Honda

Get ready to discover the most actionable end to end automation advice for some of the smartest testers on the planet. Hey, I'm Joe Calantonio, host of the Test Guild Automation Podcast and my goal is to help you succeed with creating automation awesomeness.

Hey, it's Joe and welcome to another episode of the Test Guild Automation Podcast. And today we'll be talking with Mush Honda all about AI powered tools and software testing. So we're really going to take a deep dive into a lot of these different AI technologies and how they can help your automation testing efforts. Mush is a senior engineer leader and he really leads the space in the AI and ML area especially, you know, he's a big enthusiast. He has a track record of delivering exceptional results. I believe he has over 20 years of experience in quality engineering and agile software delivery and he also possesses a diverse background encompassing healthcare, voice analytics, mobile and e commerce industries. So he knows all the different areas and he holds an AWS certification in machine learning and is a certified cloud practitioner, so really knows his stuff. He also is a chief quality architect at Catalan, which is an AI augmented test automation platform that helps teams of any size deliver better software faster. No. Really excited about this discussion. You don't want to miss it. Listen up.

This episode of the Test Guild Automation Podcast is sponsored by the Test Guild. Test Guild offers amazing partnership plans that cater to your brand awareness, lead generation and thought leadership goals. Get your products and services in front of your ideal target audience. Our satisfied clients rave about the results they've seen from partnering with us. From boosted event attendance to impressive roi. Visit our website and let's talk about how Test Guild can take your brand to the next level. Head on over to test guild.info and let's talk.

Hey Mush, welcome to the Guild.

Hey Joe, thank you for having me.

Great to have. You know, it's been a while. I think last time we spoke was like 2019. So I guess before we get into it, I think I botched your bio. So is there anything in your bio that I missed that you want the Guild to know more about?

I think the most important thing that you missed completely is that I'm a self acclaimed foodie. I love trying different cuisines from all over the world. So that's my go to these days.

Awesome. Do you have a favorite?

Yes, I actually tried Indonesian last night in fact and that was something. That was amazing. All credit to my friend here in Atlanta, sir, for making me, you know, more aware of what all we have in Atlanta for that.

Nice. My wife and I are foodies as well. We'll travel to locations just to eat. So I'll definitely have to drive to Atlanta then and check that out for sure.

Absolutely, you should.

Awesome. Very much so. You know, I thought we'd dive into. I know it's a big topic, everyone's talking about AI and machine learning, but I think you have a unique perspective in that, you know, you're an ML AI expert. You also happen to work for a vendor that has a lot of customers all over the world. So you can actually see how it's implemented, how users are using it, how they're benefiting from it. So I just wanted to get your perspective on what's your thoughts on AI and machine learning and software testing. Is it really a thing? Is it hype, where it can help folks, things like that?

So I'll be the first one to say it actually sounds like more hype, quite honestly, just because it's such a. Interesting buzzword that everybody wants to talk about. Right. So I think there's hype that precedes everything that comes up with AI. And I honestly believe, you know, one of the biggest things I observe is a lot of my fellow practitioners have, you know, the same mindset, I think, as we did maybe 15 years ago when test automation was coming onto the scene, where everybody was thinking, this is going to now replace us as people. My first sort of reaction to that is absolutely not. You know, I look at anything to do with AI and ML more as an accelerator. So the way I would say for people to think about AI is almost in the same sense of a tool that will help us get more efficient and really help us not have to worry about the mundane stuff. Right. If I were to summarize what I think of AI, it's almost like, you know, I use the analogy of Iron man and Jarvis. It's like, hey, I am in control. I am Iron man. And Jarvis is that silent partner of mine who is doing all of the heavy lifting for me and bringing things, making me more aware of what's going on in areas I may not be focusing on, but I am still in control. I am in the driver's seat.

Love it. No, I love that analogy. Another one I really like is someone says, it's like a friend, where it may give you suggestions, you may take it, you may not. And sometimes it's a bad friend, but sometimes it's a good friend. So the advice, it's up to you to take that advice, what you do with it. I guess is how it works.

Absolutely, absolutely. Absolutely. I think for me it's always been about acceleration. Right. And really, you know, being a lazy developer all throughout my career, I think that's where this comes in very handy, to be able to say, hey, I have the underlying domain expertise. Let me now show you and teach you how to get things done my way and use AI and ML, sort of in that context, to be able to say, here's the context I want you to think about only, and then go do these things or these activities for me using that context. And you know, speaking of context, that's where all of the LLMs and all of that come into play. Right. And that's sort of the new way of looking at things when it comes to testing.

Love that. So speaking of context, do you think, you know, it depends. When I go on LinkedIn, I see different trends going on in different conversations and some people are like, oh, AI machine learning should be outlawed. It's not really helping testers, it's not really a thing. And then there's other people full on like it's going to replace us. So what's the correct way to approach this? I know you said treat it like a sidekick, almost like a friend. How can it make us more efficient? And do you think it's. It's almost at a detriment if someone totally ignores the technology. Maybe not drinks the Kool Aid, but if they completely ignore it as well, I think there's probably danger there as well. What are your thoughts?

Yeah, I am sort of in between that broad spectrum, to be honest. Right. I, like any other engineer, am always skeptical as my first go to. Right. So I would say approach it in that same mindset. Right. Like I feel AI, while I've mentioned it being an accelerator. Right. It can do a whole lot of heavy lifting for you in the sense that the processing power that it contains may be the way you should think of leveraging that as a benefit. Right. So case in point, you know, I'll use an example that I've seen come up a lot. Right. We've heard different people come and talk about, hey, how do I use ChatGPT as an example? Right now OpenAI has an API version. Right. Which is great, I think from a tester's perspective. I think of it as a awesome idea sounding board. Right. Let me use something like that with the proper context and let me start thinking about how I should apply different strategies, different validation points, different criteria in the context of the software that I am responsible or helping out test and essentially raise confidence in its delivery, in its know, software lifecycle. So I think of it more as that optimizer and insight buddy, so to speak, and giving it context around what we essentially wanted. Right. So, you know, one of the things that recently came in play for me personally was, you know, just like automation. Right. I think manual testing is going to be and continue to be a critical component of the testing activity. Right. I've had discussions in the past, quite honestly, where people are surprised when I bring the mindset that I believe software testing and software development in general is a humanistic process. It is not something that is robotic. Right. We use tools to enhance that humanistic process of software delivery. And I use, and I think of machine learning as that component that will help us optimize what we do. Essentially help me get better with how accurately stories are provided to us or how accurately are reports created. How can we get value out of all of these activities that we combine and do with engineering in general and how do we make the process better? That's where I think of tooling and assign the tooling label to AI and not as a human replacer or a human bot, so to speak.

Great. So what are some contributions then that AI can play for testers? Are there a few key areas you think? And like you said, it acts like a guide. So like, what are those areas then that it could help a tester with?

Yeah. So one of the key things I think from a Catalan perspective that I can bring in. Right. Is what we've done is we've actually launched a plugin, a JIRA plugin that essentially is free to use, but what it essentially does is it lets you identify when you put in your user stories in Jira, it can actually connect and use OpenAI to create a sense of what are the different manual test cases, scenarios and steps that should be associated with validating that user story. And what we've done behind the scenes is sort of power up prompt engineering, so to speak. Right. So because of our experience with like testing and seeing how different testers actually want to validate certain user stories, we've built that information into our plugin that allows testers to quickly generate all of that. Right. And what that does is it acts as an initial baseline. Right. You know, I would not ever expect it to be 100% accurate or effective. Right. But what it does is it's, it begins as a starting point for me as a tester or as a test lead or even as a product owner. Right. To be able to say, am I giving the right level of information in my stories that will prevent unnecessary overhead for the testers, right? Where they would have to come in and keep going back and forth clarifying information. If I can get a general idea of saying, hey, am I giving enough information in my user story? So that way scenarios can be easily documented. A plugin that can do that in a matter of seconds versus a couple of hours, for example, when somebody from the team actually looks at it, processes that information and then comes back with additional question, it just helps accelerate the whole delivery process. So I think one component of that in the context of acceleration is an example. The other context of this also obviously is to say, hey, now that I see a base example of what scenarios are being driven by this user story, right? What else is missing? And me as a tester with the context of the domain of the system that I am testing, that enables me to say, well, yeah, this seems to be on the right track. Or no, not really. That means there's potentially a gap in the user story. Let me clarify the user story a bit more, right. So I can generate more accurate scenarios. Or it becomes a baseline to say, well, you know, out of the 50 that I need to do, you know, this 30, top 30 seems to be in line with what I need to do. Or maybe I need to combine these. Right. So it acts as a great sounding board and gives you an initial set of go to scenarios that you could quickly then extend out or say, well, this meets my risk assessment or risk evaluation criteria. So that's an example of where I would use that.

Nice. You know, a lot of testers are big on mind mapping. This almost sounds like a way for brainstorming where not every idea is great, but you jot it down, you see where it connects, and it's almost like a guide to get you thinking and going, oh, wait, maybe there's some risk here. I didn't even think about that. It brought up, or maybe it's not, but at least helps prompt you to start thinking. So I think it's a great partnership. Almost like you said, it's a guide and it can help you kick off some ideas, I think. For sure.

Yeah, no, absolutely. And I think, you know, just continuing on with that sort of mindset. Right. In terms of the partnering, another item I keep thinking about is I also associate these AI models or machine learning models really, to act as my bare programming partner. Right. And what I feel, it's fascinating is in the same line, we've actually also developed another feature in that's out in beta for our product, Katalon Studio. Right. It's something that we are calling as a Studio Assist. And what that essentially does is it leverages your LLM models. Again, you know, you can put in your key for your OpenAI connection, but then what it lets you do is within your idea, you can essentially type out your scenario that you want to test and it actually generates the code for your test automation script right there immediately for you. Right. And what I've been doing quite honestly is using that almost as my programming pair to be able to say, hey, I know these scenarios. Let me create and let me label this in a certain way. So that way all of my existing keywords can be leveraged. And what I found is it's almost a step up from, you know, the first leap most testers do when they first get exposed to test automation, which is record and playback. Right. As I'm sure you're seeing, a lot of the low code, no code tools that are out there for us practitioners start off by saying, oh yeah, it's so easy to use when you actually dig into it, it's nothing more than record and playback. Right, Right. And what I'm seeing is with Studio Assist, while, you know, the studio platform enables you to perform record and playback, if that's what your thing is, from my perspective, leveraging, you know, studio assist using ChatGPT or OpenAI, really what it enables you to do is it enables you to step up, away from that traditional record and playback. Because the one biggest disadvantage is everything remains static, right. Unless you go in and then say, oh, now I want to use variables, I want to use data to make my scripts more reusable, et cetera, et cetera, you're stuck, right? And then what this essentially lets you do is it lets you walk away from having to be tied in with just record and playback. And what I've also been able to do with it is be able to create scripts very, very easily, even without having the system under test up and deployed environment. So it's another way of thinking of it as almost shifting left, so to speak, where I can create my automation scripts preemptively, just like I would with my manual test cases, for example, and not have to worry about, well, I have to wait for the system to be implemented before I can essentially do a record and playback. So it's helping even like the non technical testers in my opinion, to be able to leverage and get the benefit of again, tools to be able to accelerate their life cycle. Right. It's, you know, as a practitioner I've always seen that, and I'm sure you have too, is, you know, while we talk about quality being a team sport, when it comes down to roles, it's typically people still tend to look to us to say, hey, why? You know, what can you do to make it more streamlined and give us confidence in the quality? So I think these are tools that are very, very practical for us as practitioners to be able to apply in our day to day workflows.

Yeah, I love this. You know, when I was a performance test, I used to use Loadrunner to do record and playback of me interacting with an application just to point out dynamic values. But it seems like this would automatically not only do that, but it also say, here are the dynamic values, let's parameterize it for you automatically. So it really would speed that up and I think make it more consistent where you don't miss things as well. It sounds like.

Yeah, absolutely. And while the product is in beta right now, it's available for people to download. But what we want to make sure is, you know, keep evolving it. Right. So we are always coming at it from a pure practitioner's perspective, right? Yes. All the bells and whistles and all the right keywords should exist in all of the products. Right? For sure. But I think what sets us apart is we are purely coming at it from a practitioner's perspective. Right. It's not just about using the latest buzzwords, but practically applying what can be used in today's teams, especially with the pressure of time to market and higher confidence and quality. Right. What are the practical things and what are the practical components and features that should be put into a solution that helps me as a tester be more effective and you know, put less pressure, for lack of a better word.

Absolutely. So you mentioned keyword a few times. So folks that aren't familiar with Catalan, could you give maybe a visual walk us through like what it's like to create a test and maybe how Studio Assistant helps us then?

So the best way I would say is as a commenting process. Right. When you have your script open in studio we have two modes. One is more of a non technical perspective where it looks more like traditional manual test case where you have the ability to have like multiple test steps in simple English listed out. And we have what's called as a script mode which is basically more traditional of a test automations perspective, a test automation engineer's perspective of an ide where you have your text pad, so to speak, where you can Put in your keys or your code. Right. So the way we've implemented Studio Assist is you need to be in the script view and in there just start commenting out at a step level. What are the different things that you want to apply when you're creating a workflow? So for example, if you are on an E. Comm site and you're looking to add something to your cart, for example, right. You could, in very simple terms put in those steps. Right. I typically use my Jira plugin that I was talking about a few minutes earlier and plug that into the ide, right? And then as part of the comments I can highlight that and I can say, hey, generate code for this. So what it actually does is using the context of Katalon Studio and the built in keywords, it actually uses OpenAI and then it creates those steps as coded steps that are based in groovy and that can be directly plugged in and you can start using that as part of your test. Right? So you know how I was mentioning to you earlier that it's. This is sort of like an alternative to a record and playback, a more mature way, right? What you can do in this case is you could say, well, I am looking to test a system that may not be developed yet, or a feature that's not in QA yet or in QE yet. Right? Let me look at the wireframes, let me look at the flows, let me look at what has been defined by the requirements. I can put that flow in simple English, get that preemptively converted to an automation test script, and then use my prior knowledge. Or I could use what's already been defined by the developers to say, what are my objects that align with each of those instructions? Right? So in Selenium speak, right, if you were to interact with a particular object, right, A dropdown, then what you could easily do is the actual keyword gets created for you in the script and then all you do is if you have an object repository that you've already saved, you can just simply map that to say, hey, use the name that's tied to this. It's. It may not be as clear as I'm speaking to it, but you know, on our website we've got certain examples and very simple setup instructions for it also that people could use and really do it. But what I've realized is just from a tester's perspective, things that may have taken, you know, 30 minutes, 40 minutes for me to create after an automation framework is set up, this is done in like five minutes. Or less so just being able to do a lot more with the constraint of time that VS testers face already, that's helped as a tremendous accelerator. So I've used it in that example.

Nice. So it sounds like these keywords really can help you, really help increase the understanding of test scripts, which a lot of people kind of ignore if your tests aren't readable. The more readable they are, the easier they are to maintain as well. It also sounds like once it generates like say, set a script for you that maybe does a login or enters a patient, does something like a big flow, you can automatically have it encapsulated for you as a keyword like enter patient and then you can add it as part of your. Your test suite, I guess to make it more repeatable as usable keywords.

Yes. So it actually does a couple of things. It does the keyword aspect, but it also can also do your step and flows independently if you didn't want it. So it all comes down to how you actually put the comments in there. And we've got examples on how best you can do that, but it lets you do both. In addition to that, something else that you mentioned which we've applied also is as part of this code conversion documentation and commenting is an important part as well. So this actually does that for you as part of the code conversion process as well. In addition to that, one of the other challenges we face, and this is I'm sure all of us face, is when new members join the team, right? And people who are familiar with the framework or are familiar with the context and domain of the system under test. For whatever reason, we may not be as descriptive in our commenting for the existing code. Right. And what we've also done is we've used to do assist to give another feature to it as well, where you can highlight any piece of code within Studio and you could do the exact same thing where it says, hey, explain code to me. And what it actually does is it reverses the process where it looks it up and it tells you and explains to you again with commenting to say, what is that block of code doing for you? Right. So it's almost also reverse engineering to say, well, I'm trying to learn what is happening here. What are the tests that have been probably existing before I joined the team doing. I can now highlight that and I can see the commenting and then add to the good practice of commenting code.

Oh, you can almost use it for onboarding for new testers. Hey, go through this, use this tool, have it Read it to you. Understand the code base before you dive in. That's awesome.

Exactly. And in the process, make your code better. Right. Your document.

Right, right. Love it. So you said this is in beta right now, correct?

Yes.

Have you got any feedback? Have you actually seen it in action? And it's like, oh, not only do we develop it with certain expectations, but now it's in the wild. It's actually fulfilling these expectations.

We just launched this yesterday. I believe it's beta as of yesterday. One of the things that we are excited about. Right. Is in my 20 year journey, quite honestly, I have not been a part of a team where we've built a product for the testers. So Katalon is that one place where it's awesome. Like all of our team members, we are subject matter experts on day one. So what we've been doing, quite honestly, is testing and giving user feedback as part of our delivery as well. Right. We have an initiative internally, we call it Catalon on Catalon. But we as testers say, you know, why a feature should be developed, what kind of feature should be developed, Taking the feedback from our customers and saying, hey, this makes sense. Or look, you know, as I'm going through my release or sprints, I'm giving real time feedback to our product owners to say, guys, this is not making sense, let's tweak it a certain way. And you know, honestly, for Studio Assist, the commenting piece was similarly developed. Right. It was the next stage where we said, hey, it's awesome that it can comment. Can we also make sure that we can reverse it to say, well, what happens when, you know, coding is done but no commenting is done? Can we work backwards from there?

So I always get asked this question from people too. Does the vendor use their own tool to test their own code? But it sounds like use Catalan to test Catalan, which is perfect.

Absolutely. Katalon and then some. Right. So we do Katalon and then even our partner integration tools, we are making sure we're using all of those as well and test around all of the capabilities. Right. So API, mobile, et cetera, et cetera, doing all of that, for sure.

Nice. So we spoke about generative AI copilot type functionality. These are great new features. Another term I've been hearing a lot about is autonomous testing. So is autonomous testing different? How do you define what is autonomous testing? And does Catalan have something to help with this area as well?

Right. So in the general term identification. Right. For me, autonomous means, hey, go ahead and create stuff without my input. Right. I think in general, in our industry, that term is overinflated, to be honest. What it does mean for us at Catalon, right, is I think, again, the angle of an accelerator. Right. You know, we do have autonomous test generation capabilities. We are actively working on it. Again, that is in beta right now as well. It's a close beta where we've got a smaller group of our clients that we are partnering with and working through on it with them. But in essence, what that is doing is it's a capability where we are thinking, again, from a practitioner's perspective, of what would make us be able to focus on the core value that us as human testers provide in the software development lifecycle. Right. So in that context, what we think about as autonomous test generation is the ability to look at real data as your system is being used in production or in an environment. Right. Typically, I would say production is. I go to. But looking at how your system is being used in production and then automatically creating automation scripts that reflect that usage. Right. And as part of that process, also sort of expose out two things, right? Improve and make your regression tests more realistic with areas that are being more frequently used being dynamic enough to say, every time the usage changes, your regression suite will automatically get updated with these new flows that are being used. But it also keeps you as the tester and the test team in charge. Right. So while it has the capability of creating all of that automatically, it will not do it without your input, your consent, or your criticism. So we've got the ability to be able to say, hey, let's visualize. Right. And I know you talked about mind mapping previously as well. I think that's the go to for most of us testers. Right. So we've got something that we have just released again yesterday in our latest beta release, where we have these user journey maps, which is a visual diagram of saying, hey, here's how your traffic is flowing in the system on the test. And we've identified these different workflows that should be automated. Do you want us to go ahead and automate those? And you could, you know, you could say, yes, looking at that, say, yes, this makes sense. Or you could say, no, this is not what I think it's being used for. Go ahead and regenerate. So it'll go through the process quickly over a period of time. You know, that ranges from a few hours to a few days and then growing as more data becomes available. Right. But it, it goes through that process and then it enables us to have these test cases automatically created for our consumption. Right. And then we could say, hey, I want to now use these in, you know, other environments that we are testing within. And also sort of have that as a good sense or as a partner really, to be able to say, well, now that regression testing is reflecting how the systems are being used, quite honestly, me as a tester can help my team focus on new feature development and new updates that are coming into the system and not have to do more of the mundane task of saying, okay, now that we are ready for a release, do I also have to go create and manage and upkeep my regression test? This is sort of like your coordinator, your fellow tester, or a tool in your testing process to be able to say, yeah, I already have a live upkeep of all of my regression that reflects how my system is being tested. So we've got that going as well.

So I really like this because it almost sounds like this is a very tester thing to risk. We always say focus on the risk. This can help identify risk. Because once you're in production, regardless of what your assumptions were, how you developed it, your customers may use it differently or you may not even know. Like, we have a hundred thousand tests, but you know, oh boy, only 90% of them are doing one thing and they're missing on this other 10%.

So exact.

Nice.

Yeah.

So you talked about Shift, how Studio sys, I think can help you with Shift left. It sounds like this can help you with Shift, right? Is this a feature autonomous test? Is it called something ATG or what is the this feature called in Catalan?

Yes, we are calling it autonomous Testing Test generation. Right now. I am not sure about the marketing aspect of it. You know, every time I hear terminology, my first aspect is tell me what autonomous really means. Right. And as I was alluding to you earlier, it's like autonomous in the broader sense means, oh, magic. And I know it's not magic, it's just engineering. Right. So at least for now we are calling it autonomous. I think we're going to stick with that for now.

Great. So is there anything that it can help with? Are there any limitations that you see? Well, you said, you know, it's not magic, so it may sound like magic so far, but is anything you like okay? No, I'll be upfront. It's not going to help you with these type of areas, if that's what you're thinking. Yes.

So just like everything we test. Right. I would say it's not the magic book or solution that we are looking for. That will say, hey, now we will guarantee there are no issues. No, absolutely not. Right. That's the first limitation that I think will never go away. The second one I think is because it's in beta. Quite honestly, I think what we are doing is we are still learning what are the different workflows and depending on the different complexities of different systems that are being built. Right. We are still learning a lot of saying, hey, what other ways are there that we could streamline this recording and this creation process further? Right. Right now what we have is based on limited exposure to domains. Right. So we are looking at maybe, you know, an E commerce site versus a banking solution, for example. Right. So we don't have insights yet into what are the different types of domains that this may be the most beneficial for. We're still learning and exploring that. So there's, there's a limitation where as more people sign up for beta, it may not truly apply in their context or in the complexity of their system. Right. And then of course we are still also looking at, you know, what type of systems can this support. Right. Is this more tied to just browser based APIs, you know, mobile and stuff? And right now our scope is very limited. It's going to grow over time for sure, but right now we are focused only on like the web UI aspect. So it's focusing on functional UI tests and workflows for testers. So that's the given area of at least the scope that's in play right now. And of course we are building and trying to expand that out with APIs and things coming up next.

Nice. Now I know we focus a lot on this, on testers, because this is based on testers. This podcast, however, I can see this almost helping the whole team. This is a big thing when I talk to other folks and experts is need to get the whole team involved. But sometimes it's overhead, like I don't understand Selenium, even if they're developers or I don't know what I'm looking at if I'm a business user. It seems like this can almost unify the team as well to help get everyone involved. Is that how you see it being used as well?

Absolutely, absolutely. And I think the one part that we talk about right is I think all of the features that we talked about today while we are focusing on the core activity of test creation and execution. Right. If you look at Katalon as a platform, we also have a lot of other components in that platform, which is the concept of testing and operations. We have A section that's called Test Ops, which is essentially being able to consolidate all of your reports, whether it's coming from Testng or any developer centric reports, but have a central location where as a team, you can look at the overall release readiness. As an example. Right. What are the underlying issues that are still preventing you or giving you that gauge of saying, yes, we are ready as we inch towards our deadline of release. Right. So it's definitely built as a platform that enables the entire team collectively to get together and help build that confidence in quality, really continuing to make quality as a team sport rather than the responsibility of a select few. Right. And then in the same way, apart from like being able to schedule tests and things of that nature, you have integrations into the entire lifecycle. Right. So you could go look at requirements, you could look at defects that are load. You could see all of that as a holistic dashboard, so to speak, of saying, where are we with our release? We've got integrations that go into, you know, operational environments. Right. Browser stacks, Test lambda, things of that nature are already seamlessly integrated. We have our own offering called Test Clouds as well. So it's not only just focused on test creation, but it gives you sort of like that overall ability to take testing as a holistic platform and absorb it into your overall development ecosystem. So it integrates with all of the tools that, you know, quite honestly, teams of today are using. Right. And it stems all the way from like CI CD tools, all the way to cloud execution. So we are a platform that easily comes in and works cohesively in that ecosystem.

Love it. And once again, I always do trends that I see for the upcoming year and for this year I said more and more tools becoming more platforms. And this sounds like this is the direction Catalan has gone as well.

Yep.

I know we talked about a lot of things here and sometimes people need to see it to believe it. So you did mention a free trial, I believe. Also, is there an option for a demo? And can you talk a little bit more about the free trial, how that works?

Sure, absolutely. So we actually have, for people who don't use it, We've got a 30 day all inclusive trial available. So as people download it, it starts with Studio, but then you get access to the cloud infrastructure with desktops and everything being in the cloud. So 30 day window is what we give for people to try out all of the capabilities. Right. And then of course, even if they want some demos and stuff, we've got a team through our website, people could reach out to us and we could schedule something. There's a lot of activity in the community as well as we've got a whole forum section as well where we've got active users constantly giving feedback, sharing information about how they are currently using the system, discussing ideas, alternatives. So it's a very active community. And quite honestly, the philosophy that we built Katalon with was for the testers, for the teams. So it's exciting to see that that level of engagement and activity continues on. So as people want to get more familiar with it, want to learn about it, I would say definitely head on to our website and they've got clear links where you could try all of that.

Absolutely. I also will have links to the free trial and the demo and a bunch of other information on Catalan in the show notes, so you definitely want to check that out. Okay, Mosh, before we go, is there one piece of actual advice you can give to someone to help them with their AI automation testing efforts and what's the best way to find or contact you? Sure.

So the one piece of actionable advice I always talk about is, you know, embrace change, don't be afraid of it. View as all changes as an opportunity to learn. I think being of a mindset that we say we will continuously learn, I think goes a long way. The other advice I would also give is definitely look at AI as a pro. Be skeptical, but definitely use it to streamline and make your work more efficient. Because as everybody knows, testers are definitely put under a lot of pressure all the time. So anything that we can do to offset that, definitely embrace it and try it out. The best way to reach me, I'm on LinkedIn. I think that would be my go to recommendation for people to reach out. And yeah, if we need to talk about any of the items or the capabilities we've discussed today, I'm more than happy to jump on a call and talk to people.

Thanks again for your automation awesomeness. The Linux Sentinel value we covered in this episode is head on over to testguild.com a453 and if the show has helped you in any way, why not rate it and review it in itunes? Reviews really help in the rankings of the show and I read each and every one of them. So that's it for this episode of the Test Guild Automation Podcast. I'm Joe and my mission is to help you succeed with creating end to end full stack automation awesomeness. As always, test everything and keep the good. Cheers. Hey, thanks again for listening. If you're not already part of our awesome community of 27,000 of the smartest testers, DevOps and automation professionals in the world. We'd love to have you join the fam@testguild.com and if you're in the DevOps automation software testing space or you're a test tool provider and want to offer real world value that can improve the skills or solve a problem for the guild community, I'd love to hear from you. Head ON over to testguild.info and let's make it happen.