Jeroline
All right hello everyone thank you for joining us today from different parts of the world welcome to Pcloudy's webinar on AI agents.
Jeroline
And strategies to scale and speed up digital testing we're thrilled to have you with us today I'm Jero.
Jeroline
And I'll be your host for today's webinar so before we dive in I'll cover a few housekeeping rules your lines are currently muted to ensure.
Jeroline
A smooth experience however however we encourage you to ask questions.
Jeroline
So you can submit them using the Q&A option at the bottom of your screen.
Jeroline
And we'll address them during the Q&A session at the end.
Jeroline
And also note that this webinar is being recorded and we'll be sharing the recording link after the webinar over we email.
Jeroline
So feel free to share it with your colleagues or friends who might find it valuable.
Jeroline
Now let me introduce our speaker for today Jessica Jessica is a pre-sale specialist at Pcloudy with over five years of experience in the IT industry.
Jeroline
She she specializes in test automation AI driven testing strategies.
Jeroline
And continuous testing methodologies with her expertise in modern testing Frameworks devops integration.
Jeroline
And AI power test optimization she has helped numerous organizations optimize their testing processes with Cutting Edge technology.
Jeroline
So without any further Ado let's welcome Jessica Jessica over to you.
Jessica
Thank you so much Jero for that introduction and hello every one thank you.
Jessica
So much for joining and I hope this discussion uh on AI agents.
Jessica
And strategies to scale up your digital testing would be useful to you uh.
Jessica
Right so with that let's just get started uh yeah.
Jessica
So firstly Let's uh just talking and introducing about Pcloudy.
Jessica
And who we are uh right we're a unified digital experience testing platform.
Jessica
And the core of our solution is our diverse uh device coverage where we have uh aaable device infrastructure be it your devices mobile devices browsers.
Jessica
Smart TVs watches Uh custom devices like zebra and so on where we have.
Jessica
Also secure deployment options for you be it on premise labiner box which is a quick on- premise setup private cloud.
Jessica
And public Cloud options and our solution layers comes in the form of our functional experience testing where uh you will be able to do your.
Jessica
Manual and automated functional validations for all your features in your for your builds for your applications uh be it mobile web applications hybrid native applications.
Jessica
Right and even if you're having your existing uh Frameworks like your APM.
Jessica
And selenium scripts you will be able to seamlessly integrate into our platform.
Jessica
And execute uh on our platform cross channel right and another uh solution layer is our performance experience testing which is a very exciting uh feature.
Jessica
For us uh right where we are tracking all your UI performance metrics your network.
Jessica
And apis uh your frame related response time page transition all these 60 plus um you know critical app performance related metrics.
Jessica
And we're giving those dashboards and visualizations to you and.
Jessica
Finally on the digital experience uh front where we are giving AI powered synthetic monitoring for your production related uh application where you can 24x7 uh.
Jessica
Run uh your um critical user Journeys on your production applications where anomalies.
Jessica
And trends would be detected so underg guarding all these different solution layers that you just see here.
Jessica
Right is our AI agents which drives throughout in this testing cycle.
Jessica
And uh is the underg good for our uh solution layers.
Jessica
Right be it our uh test creation agent which we'll be focusing on today along with selfhealing that's incorporated into our agent which will automatically fix.
Jessica
Your scripts orchestration agent that is that will smartly allocate test cases to your devices.
Jessica
And as well as analysis from an analysis standpoint observability.
Jessica
And impact analysis agent right uh where it will be able to detect those anomalies.
Jessica
And send alerts so that is a brief introduction into Pcloudy.
Jessica
So with that let's get into the topic for today which is AI agents.
Jessica
And before that I just want to take a quick poll.
Jessica
Right so the question is I think it should be flashing.
Jessica
Right in front of you uh right is how how often do you use AI uh.
Jessica
So here are a few options every hour every day once in a while or I don't use AI at all.
Jessica
So I'll just give you maybe a quick 30 seconds to answer that.
Jessica
And you'll see the response is.
Jessica
In yeah so I'm seeing a few responses um most of the responses I can see are within the.
Jessica
So most of the responses majority are every day and surprisingly even some are every hour.
Jessica
Right so um this merely goes to show how um Common Place AI has become for all of us.
Jessica
Right so um AI is everywhere for everyone and is used every day uh um.
Jessica
So there is no longer uh restricted technology for a research lab or Tech Giants it's for everyone uh be it all kinds of personas uh.
Jessica
Right programmers qas um teachers students CEOs right and also across Industries.
Jessica
So it Pharma banking fintech uh right across Industries we see AI being Incorporated uh.
Jessica
Right so that is why understanding its impact the challenges.
Jessica
And also specifically the opportunities it presents is very critical um both to advance your career personally.
Jessica
And also from uh making your QA and testing faster.
Jessica
Now let's talk about how AI a bit about the AI uh Evolution.
Jessica
And let's just dwell a bit on that um so firstly when we saw AI coming in it was mainly low agency it was more reactive.
Jessica
It was static It could only do simple tasks that were straightforward.
Jessica
Right uh so these are these include the early AI models that were like the deterministic chat Bots that you saw or rule-based systems that had.
Jessica
A predefined uh logic right so in the middle of um we're seeing the llm based assistance like chat Bots.
Jessica
And uh sorry like chat GPT and B uh right.
Jessica
And other geni models that are closing this Gap right.
Jessica
So they are able to perform more complex task handle Dynamic environments.
Jessica
And assist in human uh decision- making so um at the same time it's not fully closed the Gap.
Jessica
But we're moving there right and towards the future is what we see of high agency AI that will be adaptive proactive.
Jessica
And capable of completely handling complex goals and operating autonomously.
Jessica
But what we're seeing is the trend is that that agency Gap is getting closed.
Jessica
And llm models are playing a key role uh in closing that Gap.
Jessica
So uh we see that AI is transitioning from that static task based models to fully autonomous uh models that are able to complete an endtoend.
Jessica
Goal or task right on the same lines uh just this is just a timeline of how that has evolved uh.
Jessica
Right with NLP Frameworks that were more um you know text based uh.
Jessica
And it did not have contextual understanding uh moving on to conversational AA in 2017 which was intent driven it understood a bit of context.
Jessica
But it still required a lot of uh debugging um.
Jessica
And the accuracy low generative AI yes was a game changer as we all know with the launch of Chachi PT.
Jessica
And llm models and so on and AI could understand.
Jessica
And generate content intelligently right uh and it was able to give you those test script generations.
Jessica
And uh so on and finally moving in the present.
Jessica
And moving towards the future we we are in an ER era of llm powered AI agents.
Jessica
So where AI just doesn't generate content it is able to reason decide.
Jessica
And act autonomously and complete a task or a goal.
Jessica
Right end to end without human intervention so from a testing perspective this could be generating a fully functional test script.
Jessica
Right or self-healing or prioritizing and executing scripts and so on.
Jessica
So that is how we are seeing the evolution of AI agents.
Jessica
And with that let's just have a quick definition of what uh AI agents are.
Jessica
Right these are autonomous or semi-autonomous software entities right that use AI to perceive decide act.
Jessica
And Achieve goals in digital or physical environment so unlike um traditional automation which follows predefined scripts AI agents have the ability to learn from data.
Jessica
Make those decisions dynamically and accomplish the task right so it is really goal oriented in that way as we go deeper into the webinar.
Jessica
And also further into the live demo we will see how that all plays in to our solution.
Jessica
And at Pcloudy we have really been in the Forefront of this trend.
Jessica
Right in terms of making these agents um really workable in the app testing space.
Jessica
Right so um in the form of AI driven software uh augmentation.
Jessica
And really revolutionizing the way organizations can scale and optimize your testing process through.
Jessica
So it's no longer just uh you know just any other uh AI solution.
Jessica
But truly something that will be efficient for you truly that that will be something that'll make uh the life of a tester much easier uh.
Jessica
Right be it in the form of autonomous test scripting or a power defect analysis um selfhealing that's Incorporated in or smart test execution.
Jessica
Right so from an end to end goal right not no longer just giving a a script that might require uh further debugging.
Jessica
And spending time debugging uh but more over from an.
Jessica
Overall testing cycle perspective how um AI agents can truly become an efficient solution for you.
Jessica
And in that context AI agents are part of the bigger solution winning formula.
Jessica
Right forming that critical piece of the puzzle where AI agents is bringing that intelligence adaptability.
Jessica
And automation into testing uh be it in the form of uh accurate uh test case generation self-healing or intelligent test execution along with you know.
Jessica
In being implemented in an agile methodology uh tracking the coverage.
Jessica
Right increasing the you know the test coverage that you have from manual versus automation um ability to integrate into your cicd.
Jessica
And so on and of course what is again critical is having a flexible infrastructure.
Jessica
So this is really important in the sense of scalability.
Jessica
So uh AI driven test execution is really optimized for cloud-based testing across thousands of devices.
Jessica
Right so where you can run tests in parallel uh.
Jessica
Right in real devices reducing your execution time uh being able to have that comparative analysis between devices executing crossplatform um use cases Omni Channel use.
Jessica
Cases and so on and finally also having ml Predictive Analytics.
Jessica
And insights uh in terms of your apps performance as.
Jessica
Well so all of this integrate to become a winning formula for you for your.
Jessica
Testing so with all these different uh you know s um part of solution lines or solution layers that come in uh is where Pcloudy we.
Jessica
Have designed that AI agent for everything testing and and we're introducing Qpilot to you uh where it is a platform agnostic fully autonomous AI testing.
Jessica
Agent so it's gold driven in terms of Performing those testing tasks accurately.
Jessica
And it works across right be it for your mobile um uh apps your web apps.
Jessica
And soon to be AI API testing as well and it uses that agentic AI to make testing intelligent.
Jessica
And adaptive so how is it different from uh say other gen AI tools like chat GPT.
Jessica
Right so geni responds to single prompts but it may not ensure that gold completion uh.
Jessica
Right it might require further debugging you're not sure if that particular expath of the locators are correct.
Jessica
And so on whereas uh with our Qpilot agent it sets the goal.
Jessica
And plans the workflow and executes uh the task right learning.
Jessica
And giving multiple feedback loops so let's just um take a pause.
Jessica
And see how that works right so some of the uh process behind how Qpilot is working is firstly it requires simply a natural language input.
Jessica
You do not need any coding skills or there's no standard format in which you need to uh give these prompts um.
Jessica
Then it further interprets and reasons understanding what object are you referring to what action is it referring to.
Jessica
Right and it's uh and based on that it generates a script.
Jessica
Right that's the workflow generation and every step is also simultaneously being executed.
Jessica
And if it f cases any error it's giving a feedback to the workflow generation uh to make the script or the step accurate.
Jessica
So this is where the learning and course correction happens uh.
Jessica
Right multiple times iteratively uh till the step is executed successfully.
Jessica
And then you get the final outcome of a fully functional working script.
Jessica
And more from the perspective of the same workflow in terms of the agents that are happening uh working uh you know uh behind the scenes.
Jessica
Right so firstly uh the workflow agents of the expert look s.
Jessica
Right where it's a uh it goes through the entire Dom structure.
Jessica
And hierarchy right uh seeing where are those different objects.
Jessica
And elements uh trying to locate the accurate X paath in the locator.
Jessica
And uh incorporate that into the script uh second is the code generation uh which will um perform the particular action be it uh typing a.
Jessica
Text or toggling a button or selecting an option from a drop down whatever that particular action is it'll be able to generate the code for.
Jessica
It and then the step execution agent where that particular each.
Jessica
And every individual step is executed and feedback is given in case if there's any errors.
Jessica
So that um once multiple times this feedback happens between the step execution.
Jessica
And the code generation agent there's a fully functioning and working script.
Jessica
Right so this is how the agents are working um behind the scenes to ensure that you have a fully functional.
Jessica
And working script finally uh here are some more exciting features that we've incorporated into uh Qpilot since uh the.
Jessica
First time we announced Qpilot to you so first is validations.
Jessica
And assertions so uh you'll be able to validate add validation.
Jessica
And assertion steps again in a plain uh natural language input like for example validate if um add to cart button exists.
Jessica
So validate if um uh you know login button is enabled.
Jessica
So you'll be able to add some uh you know these uh validation.
Jessica
And assertion steps pretty much seamlessly uh you can also add if else conditions.
Jessica
And um you know uh weight times and much.
Jessica
More another thing you'll be able to do is a parameterization or incorporate test data into your uh use cases.
Jessica
So in case if you do not want to input static data you will.
Jessica
Also be able to give your uh test data in the form of adjacent format uh which I will be showing to you uh.
Jessica
Right so that you do not need to input uh static data uh.
Jessica
Next crossplatform test Suites so we have a powerful feature called collections uh which will enable you to look at all the different uh you know.
Jessica
Use cases from your different folders you'll be able to um create.
Jessica
And pick and choose your test cases and create a collection all.
Jessica
Right which will be crossplatform in the sense of it can be from web mobile uh Android iOS uh.
Jessica
Right and then you'll be able to give a name to your collection um.
Jessica
And select which devices you want these test cases to be executed on.
Jessica
So uh we have included collections uh crossplatform test Suites.
Jessica
And these are also can these collections can also be triggered from your cicd be it Azure devops or genkins we integrate with your cicd tools.
Jessica
As well uh next is test configuration so say for example you have already generated um a test script using Qpilot.
Jessica
And then you're getting a new build or a new URL uh.
Jessica
Right for um uh you know you do not want to regenerate again you can just go ahead.
Jessica
And um just simply modify the uh app name or the URL.
Jessica
And rerun the existing uh test case in order to not generate it again.
Jessica
So we have a test configuration option and finally we have test orchestration which will smartly allocate uh you know the sequencing.
Jessica
And execution of those multiple test cases so ensuring that you know the way your collections are executed are efficient.
Jessica
Right so be it running your full uh regressions or test suits um being able to smartly allocate that.
Jessica
So these are some of the uh few features that we have um you know to name a few uh added on since we have.
Jessica
First uh announced Qpilot now let's see more into what are the specific uh you know uh use cases where Qpilot will really help you scale.
Jessica
Up your testing and really change your testing game right.
Jessica
So now be it scaling up your functional experiences testing.
Jessica
So this can be for two scenarios say you have already your existing framework say you already have uh developed a whole bunch of um scripts.
Jessica
Right uh be it with your APM or selenium Frameworks you can leverage that.
Jessica
And Trigger your test suedes from there and we have our parallel execution agent.
Jessica
And we have selfhealing in place and as well as orchestration.
Jessica
Now the same flow can be done and be done more efficient efficiently with Qpilot sorry.
Jessica
So you'll be able to uh you know from seamlessly generate test scripts.
Jessica
Right and create your test Suites and then run uh your test cases parall on maybe 10 20 based on your concurrency there is no uh.
Jessica
You will be able to run multiple parallel uh tests.
Jessica
Right and we have selfhealing Incorporated uh in that and.
Jessica
Also test orchestration which will smartly allocate those tests across devices.
Jessica
So it will reduce that uh you know your test cycle time.
Jessica
And bring in efficiency in when you're uh running your collections or test Suites.
Jessica
And we have uh our you know the core of our solution which are which is our device browser uh infrastructure 5,000 plus device browser infra.
Jessica
Right where you can execute on that and get those Advanced automation test reports uh.
Jessica
And analytics as well so this will really help you scale your uh you know uh function experience testing.
Jessica
And what is the impact here in terms of um for you it is faster release Cycles uh with parallel execution uh lower maintenance efforts for.
Jessica
Your scripts uh efficient uh test case generation that's much faster uh.
Jessica
And accurate right you don't have to worry about debugging your scripts scalable testing because we have our cloud-based infrastructure uh.
Jessica
Right and data driven quality improvements because AI is always learning.
Jessica
And getting those feedbacks right so this is how C P it will really help you scale your functional experience.
Jessica
Testing yeah another important um aspect where Qpilot can come into play is synthetic continuous monitoring.
Jessica
Right so uh continuous monitoring of your app in production is.
Jessica
Also something um that we see a lot of Enterprise applications looking uh looking into.
Jessica
Right so for example for your critical user Journeys like for example login.
Jessica
And authentication or doing a critical transaction uh account balance inquiries.
Jessica
And these kind of um you know critical user Journeys uh can be configured to run 24x7 on our hybrid Cloud.
Jessica
Right so you can configure hundreds of parallel tests 24x7.
Jessica
And keep getting those realtime insights and metrics so that you are able to get those um you know app performance metrics your UI metrics or.
Jessica
Your uh Network and backend API performance metrics uh right you're able to get all these realtime information uh.
Jessica
Right so that uh again the impact here is you're able to detect issues uh before real users are impacted on a large scale uh.
Jessica
Right gain deep insights into the app's performance and stability.
Jessica
And maintain a seamless experience across platforms and uh devices.
Jessica
So this is another way where Qpilot can really help make your synthetic monitoring uh much more efficient.
Jessica
So with that uh I think we are done with the slides.
Jessica
Let's quickly go into a demo of Qpilot and we'll see it real time in action here.
Jessica
So I'm just moving to the tool give me a.
Jessica
Yeah so this is our uh this is the Pcloudy uh landing page.
Jessica
Right uh that you see here and as I was telling you this is our UniFi digital experience platform here you're able to see our new.
Jessica
Releases on Qpilot and app performance testing as well and uh you know our diverse device browser infrastructure along with all the various Integrations.
Jessica
And capabilities that we have uh so with that I'll just go quickly go into the devices section.
Jessica
So the devices that you see here are all real devices that are hosted in our data centers.
Jessica
Right so we do not support simulators or ulators these are um you you know real devices that are hosted in our um Global device infrastructure.
Jessica
So you'll be able to filter based on the OS version.
Jessica
So we have the latest um OS versions be it Android 15 or the latest iOS version with uh 18.3 uh.
Jessica
Right and so on you can also choose from your oems we have all the different devices uh Apple Google Samsung uh.
Jessica
Right and screen sizes Network as well so in case if you need uh any Sim based devices for your testing you'll be able to do.
Jessica
That as well and uh finally also based on location.
Jessica
So we have data centers multiple locations in India the US Singapore.
Jessica
And also our newly launched um data center in the UAE uh.
Jessica
Right so with that introduction let me just quickly go ahead.
Jessica
And connect to a device so once you click on connect uh.
Jessica
Right um You will have full access of the device uh with you.
Jessica
And prior to testing let me just uh while the device session is getting established let me just show you another section where you need to.
Jessica
Ensure your app is uploaded right so in the my data section ensure that you go ahead.
Jessica
And uh browse through your machine and upload your uh APK or your IPA bills.
Jessica
Right and once that's done uh I'm ready to begin my testing.
Jessica
So um this is in the manual mode so I just want to highlight a few features before we get into Qpilot uh mode you'll be.
Jessica
Able to see the logs take screenshots analyze performance metrics like your CPU memory take screenshots uh log bugs in your jira or your any bug.
Jessica
Tracking tool right and we also support specialized features like um image injection.
Jessica
And um for scanning QR codes and checks and so on.
Jessica
And also Biometrics simulating uh Network for different uh Network speeds geolocation testing.
Jessica
And so on so if you're interested to know any of these features you can of course uh look at our customized demos or you can.
Jessica
Reach out to us for a customized demo uh right uh.
Jessica
But I will now go into the Qpilot mode so.
Jessica
Now when we're in this mode the only input that is required is a plain natural language.
Jessica
So I'm going ahead and selecting our sample.
Jessica
Application giving the banking sorry the package name and the app activity name I'm giving a logical feature name.
Jessica
And I'm giving a test description right so this test description can just be in plain natural language as you can see what I'm going to.
Jessica
Paste here right is a plain simple easy to understand natural language.
Jessica
So there's no one uh format as such that it needs to be in you can uh.
Jessica
Also incorporate test data if you do not want to uh add a static data.
Jessica
Right so let me just show you how that works as.
Jessica
Well so in this case if I do not want to enter the actual name uh I can input test data in the form of ajacent.
Jessica
Format so it's again pretty much straightforward click on ADD test data give you all your uh the variable name.
Jessica
And the corresponding names in this format and click ADD test data.
Jessica
So once you do this uh you can see and you can just uh call that variable in that particular step.
Jessica
And uh the name will be incorporated uh in its place.
Jessica
Right so with that I can just save it select the folder that I want the test case to be saved in.
Jessica
And give a test case name yeah now I'm going to go ahead.
Jessica
And click on generate so as soon as I click on generate what is happening is firstly uh the app will get launched uh the based.
Jessica
On the details that we've inputed here right with our app package.
Jessica
And app activity name the app that we have selected will get launched.
Jessica
So yes the app has gotten launched and one by one the AI will understand uh the steps that we've given here all.
Jessica
Right the natural language test case scenario that we've given here.
Jessica
Right so for example enter demo in username field uh all the agents that I was discussing.
Jessica
Right is coming into play here so with um the username field that's specified it's going to go through the entire uh Dom structure hierarchy.
Jessica
And uh identify accurately what the username field right with the expath.
Jessica
And the locators so that is the added Advantage here in the sense that you do not need to go.
Jessica
And uh you know use your uh APM desktop or any other object spy in order to get those expats.
Jessica
And the locators uh it is able to fetch them seamlessly.
Jessica
And it's able to enter uh do the action which is enter a particular text enter demo.
Jessica
Right so this does take you know a few seconds of time.
Jessica
But the biggest Advantage here is that you get a fully functional working script that I was mentioning.
Jessica
Right uh so uh that is the added advantage that you get here in the sense that maybe there's a minute or two for a test.
Jessica
Case to get generated uh but uh the script that is once it's generated is fully accurate.
Jessica
And you can directly trigger it there is no need to uh sit.
Jessica
And debug uh your script and so on and another thing I want to highlight here is that it pretty much Works seamlessly the same way.
Jessica
For browser where instead of these details you will be giving your uh URL.
Jessica
And again you just give the same natural language description uh.
Jessica
Right and it will it will launch the browser launch the URL.
Jessica
And one by one each and every one of these uh steps will get uh executed.
Jessica
Right so it's pretty much the same way for a browser as.
Jessica
Well and another thing I wanted to highlight is once um your test scripts are generated you will be redirected to a dashboard called The Qpilot.
Jessica
Dashboard um where you will be able to see all your test scripts in a folder hierarchy.
Jessica
Right so you'll be able to organize your test cases uh.
Jessica
Right rename them uh and uh you'll be able to execute them individually or.
Jessica
Also group them into collections uh and run them as a collection as.
Jessica
Well yeah so I think where are we we are.
Jessica
Now logged in right there are just a few more steps to be done uh.
Jessica
Right so we're just going ahead and performing a simple transaction here uh where uh this is our sample banking app.
Jessica
And we have logged in and we're just doing a fund transfer.
Jessica
Right uh so as I was mentioning you're able to see.
Jessica
Right that for the feedback between the code generation agent.
Jessica
And the workflow execution agent is happening right now in the back end.
Jessica
So for every step uh you are able to see um uh it's able to generate a script execute it.
Jessica
And validate that whether the step was successful or not.
Jessica
So in this case you can see that it has taken the data from our test data input.
Jessica
And inputed that particular recipient uh name in that particular field.
Jessica
Right um and then yeah we just have a couple of more steps to go with entering the amount.
Jessica
And clicking on transfer yeah so I think uh we're pretty much done the test steps are completed.
Jessica
And you can see now that there's a popup that is saying that our script has been generated successfully.
Jessica
And it's redirecting us uh to the Qpilot dashboard this is the Qpilot dashboard that I was talking about here in the all Test Section is.
Jessica
Where you'll be able to see all the test cases that you've created uh.
Jessica
Right uh and in a folder format and you'll be able to um uh you know um organize it in the way you want it to.
Jessica
Be so here is uh the script we have created I'm just clicking on script.
Jessica
And as you can see here is our fully functional generated script.
Jessica
Right you can generate it in multiple languages JavaScript python Java.
Jessica
Right we have multiple languages that are there uh you'll be able to see your test data over here.
Jessica
Right um You can also go ahead and edit your test configuration.
Jessica
So this is what I was mentioning um as one of the uh addon features that we have uh with Qpilot where you in case you.
Jessica
Get a new build you do not need to go ahead.
Jessica
And regenerate the whole test case again uh you can just simply uh select the particular new build uh.
Jessica
And uh from the drop down right and save that.
Jessica
And you can also in case uh if there's a change in app activity package name you can.
Jessica
Also change that or the feature name and go ahead.
Jessica
And save it you can also go ahead and run an individual test fit all you need to do is click on run.
Jessica
And go ahead and select the particular um devices that that you want it to run on from the drop down.
Jessica
So here is the list of all the available devices you can choose uh for it to run on.
Jessica
But I want to go ahead and before running it here itself uh show the common scenario where it will be triggering multiple test cases uh.
Jessica
At once so let's just go to the collection feature yeah.
Jessica
So this is the collection feature that is a really unique.
Jessica
And differentiating feature here at Pcloudy which is truly where you'll be able to achieve true crossplatform testing.
Jessica
Right be it your uh web test cases uh right your iOS your your Android use cases you'll be able to trigger them all together in.
Jessica
One shot so I'm just going to go ahead and add a few.
Jessica
Um Android use cases and iOS use cases and web based use cases.
Jessica
Right and you can give a name to your collection give a description if needed.
Jessica
And you can give the project name so that your report is generated specifically for this collection along with a build name uh.
Jessica
Right and here are all the s uh use cases we have selected from here.
Jessica
And then you can go ahead and select which devices you wanted to execute on.
Jessica
So for example I'm just selecting the version range so for example here version 13 to 15.
Jessica
And uh the OEM I've selected is Samsung you can just select uh the your Preferred Choice uh I'm doing the same for iOS.
Jessica
And along with the number of devices as well you can choose.
Jessica
And I'm doing the same for browsers so now once we have given uh the device count.
Jessica
And the test cases that we want uh we can go ahead.
Jessica
And save our collection so this is redirecting us to the all collections.
Jessica
So here uh you can directly trigger the run from here or we.
Jessica
Also have so I'm going to be triggering it here from Pcloudy's UI.
Jessica
But you can also trigger it from your cicd as.
Jessica
Well we as I mentioned we integrate seamlessly with uh cicd tools uh.
Jessica
So I'm just going to go ahead and click on run on a particular collection.
Jessica
And as you can see the collection execution has started.
Jessica
So let me just go back to our active sessions to see if the triggers have gotten if the triggers have begun.
Jessica
And also what another thing you can do is you can go ahead.
Jessica
And monitor the reports well so you'll be able to see your Progressive report.
Jessica
So even before the uh execution completes you will be able to uh track the reporting for your use cases that are.
Jessica
Triggered so here as you can see we've gotten uh some automation sessions that are uh ongoing we can go ahead.
Jessica
And click on the live view.
Jessica
Here so here basically you'll be able to see what uh whatever executions that you've triggered you'll be able to see it live there are.
Jessica
Also a few options here where you you can zoom in zoom out.
Jessica
So here you can see the transaction is happening now uh.
Jessica
Right the automation is ongoing uh in case if you want to do a manual uh intervention you can do that uh else there is pretty.
Jessica
Much nothing you need you need to do from your end The automat automation will happen pretty much seamlessly uh.
Jessica
Right and I think flow is almost complete and once it's done the session will get released um.
Jessica
And subsequently all the other while all the other sessions are.
Jessica
Also happening right now in parallel so the execution is done.
Jessica
So let me also go to the reports section and uh show you how the reports will look like.
Jessica
So here is the reports that we've triggered.
Jessica
So you'll be able to see the breakup of all your test cases.
Jessica
Right how many are completed how many are failed in progress.
Jessica
And so on so you'll be able to see all of these for every individual test case you will be able to get this report.
Jessica
And moving on into a sample report and showing you how a sample report looks like for an individual test case uh you will be able.
Jessica
To see the text logs and the output of each.
Jessica
And every step as you can see here with the output.
Jessica
And in case if you've enabled for screenshots even the screenshots will be visible here like you can see over here.
Jessica
Right um also you'll be able to see the device logs uh.
Jessica
And also you can download that as well the APM logs.
Jessica
And that's also downloadable uh along with the app profiling which will give you the CPU.
Jessica
And the memory details and along with this you'll be able to see the session the video of the entire session that has taken place.
Jessica
So this is the kind of report you'll get for every uh individual test case uh.
Jessica
Right and uh this is how Qpilot as I was mentioning can really help you scale up your functional testing.
Jessica
So uh be it from so I have shown you from the start of creating a script uh from exe from generating.
Jessica
And that script and editing it if required uh in terms of the test configuration uh running an individual script or running it at scale through.
Jessica
Collection so this is how Qpilot can really uh uh revolutionize the way you do.
Jessica
And scale your functional testing so uh I hope this demo has given BR you a glimpse into our capabilities.
Jessica
And of course if you're interested for a more customized or a more detailed discussion or a demo please do reach out to us uh we.
Jessica
Would be happy to connect with you uh so with that I I would move to the Q&A session.
Jessica
So if there are any questions I will be taking that up.
Jessica
Now yeah let me just see the chat.
Jessica
Um yeah okay I've gotten doesn't support UI testing given the designs over apps like in figma.
Jessica
So um as of now uh we mainly do it for builds like APK.
Jessica
And IPA builds uh but this is interesting we can have a further discussion as of.
Jessica
Now it's not um not for figma um okay here's another interesting question can we automate flutter apps to using Pcloudy yes that is a yes.
Jessica
Uh you will be able to um uh automate for flutter apps we have done this before.
Jessica
And achieved uh use cases successfully so it again pretty much works the same way you upload your uh build uh.
Jessica
And give the natural language inputs and you the test scripts are generated irrespective of the framework.
Jessica
Right so even in this case of uh IR respect of the technology.
Jessica
So here if your app is built on flatter we will be able to support Qpilot for flatter as.
Jessica
Well can we do it for different business apps yes there is no limitation in terms of uh you know um in terms of the sector.
Jessica
Of the app so be it for banking or e-commerce there is no limitation.
Jessica
So as long as you have the bills uh you can test for multiple applications uh you just need to upload it in the my data.
Jessica
Section that I showed you and you you will be able to do it for different business apps there is no restriction there another question that.
Jessica
I see is can we extract the code and use in local selenium yes you'll be able to extract the xarts there might be a few.
Jessica
Tweaks you need to do there uh but uh you can extract the expas the locators.
Jessica
And the code uh from uh you know the Qpilot generated one.
Jessica
And reuse for selum but there might be a few tweaks you need to do here.
Jessica
So uh yeah the selfhealing yes definitely that is incorporated uh with Qpilot.
Jessica
Right uh so we can definitely have a detailed uh discussion on self-healing.
Jessica
And reach out uh to your specific use case on self-healing uh we can definitely have uh another round of discussion on the self-healing part.
Jessica
But just to show you how that works is if there's any change in your new build all you need to do is regenerate the Qpilot.
Jessica
Run uh right and it will generate that fresh script for you uh.
Jessica
So you do not need to do those fixes because Qpilot is able to seamlessly in a matter of minutes just generate the new script.
Jessica
So you will not um uh it will automatically be fixed in that.
Jessica
Sense can we integrate with text execution tools like jira yes we integrate with uh our HPM uh yes we will be able to inte integrate.
Jessica
With these tools we already have integration with jira uh.
Jessica
Right so so uh you can for example logging a bug uh.
Jessica
And so on you can simply just log it from Pcloudy.
Jessica
And it will get logged into your.
Jessica
J can you create automated where two devices communicate with each other uh.
Jessica
So in the sense I would I need to understand the specifics of the use case we can definitely connect uh further to see how that.
Jessica
The feasibility of your particular use case but definitely we will be able to um generate uh use cases simultaneously.
Jessica
So I was showing it to you one use case being generated at a time.
Jessica
But based on your concurrency not just two you can uh do three or four test cases at one time that is getting uh generated you.
Jessica
Can do that can we generate test cases from backlog as.
Jessica
Well so again uh the in Qpilot mode it requires the natural language description.
Jessica
So as long as you're able to provide that you can uh you can generate the use cases uh seamlessly.
Jessica
Okay can we do performance and load testing using Qpilot.
Jessica
So for that we have for performance testing uh not load testing.
Jessica
But uh load testing you can trigger your jmeter scripts.
Jessica
Right uh whatever jmeter or the load uh that you're going to trigger on the server you can trigger it from your jmeter uh.
Jessica
And we have the we have a performance uh engineering uh module which is not the Qpilot module that's a separate seate uh solution entirely which.
Jessica
Is another solution layer of Qpilot uh sorry of Pcloudy where you will be able to do a comparative analysis of a performance report uh of.
Jessica
Your application without the load and versus with the load.
Jessica
So for example we are tracking 60 plus metrics uh.
Jessica
Right and you will be able to do a comparative uh uh you know you'll have the comparative reports of um a report with the load.
Jessica
And without the load so you'll be able to do a comparative app performance testing or while you're triggering your load scripts from uh your load.
Jessica
Testing from jmeter scripts so I.
Jessica
Think you will be able to duplicate uh test cases uh yogiraj uh yeah can we copy one uh test case.
Jessica
And use in another test case you can uh duplicate it uh that is possible yes can we use it for automating the apis yes that.
Jessica
Is uh in pipeline as of now also we do have an API test where you'll be able to give curl API uh data.
Jessica
Right so you can just give the curl API as a step.
Jessica
So say for example your app is um in a in a very early development phase.
Jessica
And you do not have the login button or any particular um for the UI to trigger a particular action you can give the Cur apis.
Jessica
And uh you know um execute that particular step and uh API testing is.
Jessica
Also in the pipeline uh does it have Aid driven missing test case detection I would need to know more about uh your particular use case.
Jessica
Um you know what uh uh you know the specific use case of what you're um meaning here we can definitely have another discussion.
Jessica
So as of now to detect the Miss Missing use cases um maybe not directly maybe this might have to be a customized thing uh we.
Jessica
Can have a second round of discussion for sure on this thank.
Jessica
You so I think we have pretty much covered any.
Jessica
Uh uh most of the testing yeah okay vulnerability and security testing uh no as of.
Jessica
Now we do not support security testing I would uh need to check that.
Jessica
And get back but as of now we do not support security or vulnerability.
Jessica
Testing okay I think it's uh most of the questions have been covered thank you.
Jessica
So much for all your questions I mean it has been it's been really positive to hear all these questions.
Jessica
And to know that the session has been um in some way uh you know inspired you to think of how you can incorporate AI into.
Jessica
Your testing and Qpilot um to help you with your testing Journey.
Jessica
So thank you so much for these responses that's really been great.
Jeroline
Yeah uh so that brings us to the end of today's session a big thank you uh to our speaker Jessica for sharing such valuable insights.
Jeroline
And to all of you who have been an Engaged.
Jeroline
And interactive audience um if we couldn't get to your question or if you have more questions don't worry you can always reach out to us.
Jeroline
At info@pcloudy.com or connect with us over LinkedIn and.
Jeroline
Also just a quick reminders uh like I said before this session is being recorded.
Jeroline
And we'll be sharing the link with you right after this webinar over email.
Jeroline
And feel free to revisit this discussion try out Qpilot.
Jeroline
And also share it with your colleagues as well and stay tuned for our upcoming webinars.
Jeroline
And be sure to follow us on LinkedIn because that's where you'll find more updates um on our AI driven testing capabilities Automation.
Jeroline
And Industry best practices so until next time happy testing.
Jeroline
And have a great day thank you thank you.
