Transforming DevOps on AWS: Leveraging Generative AI for Next-Gen Automation and Efficiency – Predict 2025

Hi, so my name is Ape Shaw. I'm the senior partner solution architect at AWS, and today I'm going to use, uh, what is quite hot in the industry, which is how we can leverage the gen AI for the next generation automation and efficiency. Um, and the answer to this problem is Amazon Queue Developer, um, which is, uh, Amazon's product, uh, which helps, uh, the developers, uh, to, uh, do a lot of things.

So, which we will go through in the details. So what is the agenda for today? So we are going to start talking about why generative ai, uh, what is the need for the customers, uh, give you the overview of the service, what is the key features about the service, uh, some of our customers who have used this services, and how you can get started, uh, with the features of Amazon Queue developers so that this becomes part of your everyday work.

So, um, we did a lot of analysis about this before launching this service, and what we found was that, um, a median developer spends only an R writing the code that is developing new features and, uh, getting something, uh, straight to the business, what they're looking for, rest of the seven hours, he is either debugging the application, he is writing the comments onto the code, he is doing the unit testing or operation support and so on, which constitute a lot of time. Yeah. So if you take five hours in a week, that is significantly low, um, uh, effort, which has been put into to develop the new features and new business requirement.

So this is something which is quite important to understand because that's something which is what we are trying to improve upon. So what is the key needs for innovating faster? We want to build faster, we want to increase the velocity so that in a sprint we can release more things to our customers rather than spending the time on to the undifferentiated works, which is required as part of developing the software.

Yeah. Let the machine do the work, what is not fundamentally, uh, which, which can be automated, whereas what requires a new features, which is where we want new developers efforts to be put onto. Yep.

Uh, a lot of, uh, time goes into the operations. So there is an efficient way to manage and optimize the AWS cloud environment. Um, and we will go through into that as well.

Uh, there's a lot of transformation things. So basically, uh, you have a Java application, which is running on Java seven, Java eight, Java nine. You need to upgrade those two, Java 10, 11, 17, um, uh, and so on.

Because there is a new fixes, there is a new security vulnerabilities. Your security team is behind you a lot of this times, which builds up into the developer's backlog and enhance into the product backlog because the product needs to be kept up to date. Yeah.

As well as you want to do migration from t net framework to t net core. Yeah. Uh, which is where you can run the Windows software onto the Linux platform so that you can get the business benefits, uh, get the license freedom, and you can do the ization and all the benefits which comes with it.

And last but not least, but with the ai, everything is data. Yeah. So the more precise your data is, the more better you build your analytics solutions, the better the AI and ML solutions will be.

So how we can leverage Amazon queue developer to be faster in terms of developing our AI and ML solutions when we have the data with us. So if we go into that, then, uh, what, as we discussed that Amazon Queue developer helps with the quality, uh, it helps with the efficiency, it helps with the speed and how it does it, we will go through that into the detail. It has, it is built upon the generative AI and helps, hence it helps down to reduce the cost, basically.

Uh, because if the efficiency highs, if you're delivering more features as to what the business needs, then obviously the cost will be less. So there is a tremendous amount of, uh, uh, what we say the enthusiasm about saying, Hey, you know, uh, we should be using the generative ai, which will transform and which will power our business. Um, but lot of companies have, one of the thing is that, hey, but what happens about the, um, uh, my particular, uh, security framework will, my code will be used to train, uh, the, um, uh, Amazon queue developers LLM models.

The answer to that one is no. And the reason for that to make it upfront was that we make sure that we don't use the customer's data to train those LLM models. Those LLM models are already being trained, and they, we are using your data just to generate the core and provide it to you basically.

So there is nothing that this trains our LLM models. So that's something which I wanted to put forth thing into the rest. So there is a new experience altogether.

So you are, whether you are in the ID where you are, uh, trying to, um, um, do the coding, whether you are doing the mass transformation. So like I was saying, that you might have a lot of applications which you want to transform, move from dotnet framework to dotnet, uh, core, you want to boost the productivity, you want to have the content creation, which you want to do. You want to see the insights, and you want to have the creativity.

All those things can be powered with the generative ai. So what is the fundamental, um, uh, thing which we are talking about? So, uh, a customer comes to us and say, Hey, you know, we have a lot of systems.

We have thousands of applications, which has been running on true, and there's a lot of core which has been written on true, but then every time we try to build something, it takes a similar amount of effort because we cannot learn from the systems what we have built, and we cannot quickly convert them into the next level of features. What we want to develop onto, so this is where Amazon queue is differentiator. Uh, we learn from the knowledge of your company, your code, your systems, uh, and we, it's not used to train the LLM models, as I said before.

And if, when you ask the question saying, Hey, you know, I want to develop the new feature, it based upon your company's core and systems, it develops the new functions, which is required into your, uh, preferred id, which we have been integrating with. It's available wherever you work. So it's either you are in the id, whether you are into the AWS consoles, whether you are in a particular service, wherever you want, uh, the service to be available.

It's available with you in order to do the development and enhancement. Um, there are superior generative AI performance on tasks. So we have specifically integrated with lots and lots of knowledge base, which we have got at AWS so that this performs better, and it gives you the exact content, what you're looking for, the integration.

So for example, if you want to integrate with the F three, what is the SCDK and SDKs, which we have available as part of the AWS, which you can integrate with. Uh, so what does the stack looks like? So at the bottom of the stack, as you can see, there is an infrastructure.

Um, those infrastructure is our core. Uh, if you say EC2 instances, uh, which has been used, which are specifically used through train the model, those are the foundation models on which training and influence happens. On top of that, we have got the bad drug, uh, which helps with the, uh, guardrails and all the customization capabilities, what is required to power your business, uh, and, uh, uh, the LLM models, uh, so that you can access whatever LLM models, which you want to integrate with and so on.

And then on top of that, we have used, uh, certain applications, which can be used around, so you might have heard about Amazon Queue business. Uh, we will get into the retail as to what is the use of that Amazon Queue developer, which we are talking about right now, um, which helps the developers to carry out their development, operational security task into the code. We do have Amazon queue in QuickSight, so user can type in the queries into the natural language, and then Amazon queue in Connect as well.

So if the customer is calling, uh, into your data set, into, into the call centers and want to, uh, and the customer care person wants to know the specific information, then Amazon Queue can help with that knowledge base as well, with respect to quite optimizing and telling the consistent message back to your customers. So it's, Amazon Queue is powered by the ai, um, as we were talking about. So there are two kind of Amazon queue developers, Amazon Queue developer, and Amazon Queue business.

So Amazon Queue business is for every employee, uh, who are using the things like Teams or Slack or, uh, uh, you are, you might have a knowledge base, which is your SharePoint. Uh, you might have a lot of Outlook application, Microsoft, uh, which can be integrated, so that, which can be used by the business, uh, team to derive the better, um, uh, optimized information to put that into the task. And then the other side, we have the Amazon Q developer, which is for the developers and data scientists and IT professionals basically.

So that is something which we are going to talk in detail today, not Amazon Q Business. So a Q developer, as it says that it helps the developers and IT professionals build the software faster. We want more accurate coding recommendation.

Um, agents can autonomously help you implement feature If you want to say, Hey, you know, this code is very complex and someone has returned it, can you please add the code to, uh, the comments to the code? Uh, it'll add you the Java comments or T net comments or JavaScript comments is what you want. And if you, if you can, it can also say saying, Hey, you know, the code is very complex, uh, which you have written, so you can re, you can ask it to refactor the code, uh, as well as it performs the software upgrade as we were talking about.

Um, and also Amazon Queue, as you can, um, understand it, that every other knowledge which we have acquired over the years has been fit into the Amazon queue. So anything which you require the information about the Amazon AWS, you have that information available. And obviously, um, security is Job Zero at Amazon, so we never start with anything which is not built with security and privacy.

Um, and that's, that's most of our customers ask as well. So, uh, wherever you are working. So it's available in Amazon queue.

So nowadays when you logging onto the console, you'll be able to see the Amazon queue, um, which is on the right side of your, uh, uh, of your screen. Uh, if you click on that, and if you ask any questions, that's something which has been powered by the Amazon queue, and it can answer any questions, what you want to know, say how much you have spent, and saying, what is the issue with my service? And so on.

You can get the information. You can also use it into the ID integrated development environments, so like Visual Studio or JetBrains, if you're using on two, you can integrate that and start using straight thumb Amazon Q developers, um, AWS documentation it has been trained onto. So anything which you require with respect to the AWS, it provides the information about that as well.

Uh, same as I was talking about, you can integrate with Slack and Teams as well, and it's available on the mobile application as well. One recent thing, uh, which we have partnered with is our partner GitLab. So GitLab now has got the GitLab deal, which has been powered with the chain ai.

Um, and it's using the Amazon Q developer at the back of it. So, which is something fundamentally allows you, if you're using GitLab quite heavily, your all the source scores are into the GitLab. Uh, you can leverage Amazon queue straight away by using the GitLab over there.

And Amazon Queue developer is been recognized, um, as the leader. So in the AI code assistant, so we are in the leader's quadrant of the 2024. So as you can see that, um, uh, it's, it has got, uh, all the capabilities, what is required, uh, to match up those things.

So, um, as I was talking about, build faster, operate at scale, transform the workloads and leverage data ai, this is what it makes the things, deliver new features to the company and make the innovation faster for the developers rather than the mundane task which they need to go through. So we'll dive into the deeper, so, uh, build faster. How does the build faster happens?

So where are the developers spending the time? So they want to explore, uh, saying, Hey, you know, I want to build a website and in order to build a website, what all things I will need to do. Uh, I want to create the software.

How do I create the software I want to test and secure review and deploy? And then the last but not least, maintain, transform and modernize. So this is usually the time where the developers are spending it, and we'll go through, into the details by going through how in every part of the SPLC, this tool is helping you out.

Um, so let's see a demo. So as you can see, this is your id. Um, you have connected with the A WSQ developers, and you start asking the question in the consulting, Hey, you know, I want to develop a web app, how should I do it?

And it gives you the answer. Um, so that's the first exploration part of which services I should use onto the AWS in order to develop the application. It can help you with the service selection, when should you use what service?

And so on. So that's basically the exploration part before any development activities happens. Then it comes down to the creating, generating the code.

So as you can see, uh, you can just write the comment and say, Hey, I want to write a function to square a number. You connect to the Amazon queue developers in the browser. This is the visual studio.

Um, and once it's been connected, you can do all your merging. Um, the, the release of the software writing, the unique case test cases, um, writing the, uh, functions as you can see that the function is getting return, you just write the comment and you just accept those code. So the co the developers can accept the core, or if there are sometimes multiple options available, it shows to the developer which part of the function you want to use it as well.

Um, and it's, it's significantly, um, uh, what we say the, the faster way of developing the software. Even if you want to edit it, you can edit it. Um, once the base code has been available to your business specific domains, which, which LMS might not have been trained with.

So once it has been trained, um, there are, there are quite a lot of, as I was saying, the, our customers asking, saying, Hey, you know, but, uh, it's great if I want to write something completely from scratch, but my organization has got a lot of, um, source score already available. And most of the time it's the integration of, uh, one application to another application and how to do that integration and so on, which is taking the time. So what the answer to that is that you have the private repository, uh, on which the private repository you want Amazon queue developers to, uh, integrate with.

And using the Amazon queue developers, it can recommend you when you ask in the natural language saying, Hey, you know, can you please, uh, give me the list of, um, uh, unassigned food deliveries, uh, which which is around the driver's current location so that we can take the better, um, uh, we can develop this feature quickly and deliver it. Now, this is something which is, you already know the driver's current location. There is already the code, which has been written down, and you want to write a new feature, which is to say unassigned for deliveries, uh, around that.

So it leverages it, understand your code, it find out as to how to retrieve the driver's current location and based on the driver's current location that it tries to retrieve the list of the unassigned full. So it understand your code, it makes, it makes, it's not a RIC code recommendation, which comes down. It comes down to the customized code recommendation, which is what is required by your organization.

Um, there are advanced features, which is available as well. So, uh, q can write, uh, um, uh, q can write, and it can provide you the features with respect to how you can document, uh, the code. So as you can see over here, uh, there is a document you want to create a read me file, uh, read me file for the whole project as to how to use it.

And in no time, it'll scan through the source code, it'll create the knowledge graph, it'll summarize the source file, and it'll generate the documentation. Um, uh, previously this used to take, if you, if you are, I remember in the projects where I was doing the development, and if the read me file was not there, and if the developer has left, someone will come back and say, Hey, you know, I, in order to create this, uh, read me file, which you're asking, which is required for every application, it'll take me x amount of days or sometimes weeks, uh, in order to generate this is just now available at click of what is needed. Yeah.

And it is formatted, it is ready for checked in, um, and it is ready for review and then merge and release to the production as well. So this is all part of the integration. What you can see too, testing.

So from in the past unit test was one of the thing which we were putting into the estimation, and then we say, Hey, you know, I have written the code, which is working, but you know, we cannot still release because there are no unit test cases, or there are no, um, um, uh, the, uh, the proper testing which has been carried out onto this code. So now we can also support writing of the unit testing. So the agent supports that saying, Hey, you want to write the test?

So here is your clause and you want to write the testing of this methods, then we can support that as well. And it can create the test, uh, the unit test cases for every function, which you have been using it. So as you can see onto my screen, which has been showing around saying, Hey, you have four functions, and we have generated all the test cases, developers accepted it and say, Hey, come on to build and execute now.

So it is building and it is, uh, uh, it can deploy. As you can see, it's a MA one clean verify, which is happening around it's building and executing into the browsers. And then from there, it can take around in order to your ci cd pipeline to deploy to the, uh, uh, into whatever environments which you want to deploy onto security.

So one of the major thing which was coming around is, hey, great, we develop new feature, we, uh, fix the test issues, but now we want to make sure that, uh, security wise, it is, um, uh, it is quite secure. So it's the great feature about this, yes, it does the realtime check, it provides you any vulnerabilities and not only provide it with the vulnerabilities, it provides you how to fix it as well. So for example, if you're using the libraries, which are quite legacy libraries, then it can help you to upgrade those libraries and it can fix those libraries, uh, in no time as well.

And it categorizes, as you can see on the left side of the screen, that it categorizes critical, high, medium, low, or if it's just for information as well. So you have all the information which is required in order to make the features, uh, the development quicker. Um, the last but not least is review and deploy.

So we all know that once it's been, uh, once the feature has been developed by the developer, we want someone to be reviewing that code. And once it's been reviewed, then the, as part of the merge and released, then the review happens. Now, this part, um, I have seen in the company, there was a lot of debate and discussion saying, Hey, though this variable name is right and this variable name is not right, and there was a lot of friction between the developers, this developer doesn't like me, and all those things, we leave that to the system now.

So the system will say, Hey, you know, that based upon, um, the review which I've carried out, first of all, it's consistent review and based upon the review, which I've carried out, uh, I think these are the issues which needs to be fixed upon. Um, so that's something which gets reduced down in minutes to hours basically, rather than in dates. Uh, and it's a consistency of which with which the code is getting reviewed.

So the great questions which customers always ask us is saying, so what is the metrics, uh, which I should measure? Uh, because, uh, previously we, we used to, our developer used to take, say, 10 days to develop a feature. Uh, is it going to be all of a sudden it's a one day?

Uh, there is a bit of a learning curve as to how to, how to prompt, uh, the Amazon queue developer. So over the period of time, you will have the metrics, uh, which will show you that, uh, how significantly the impact which has made, uh, by using Amazon Queue developer. So we recommend these four metrics.

One is the report time saving across a range of tasks, what the developers was doing before, uh, ion acceptance rate. So basically the what is your acceptance rate, which looks like, so Amazon queue developer in your ID has recommended the code, and is it the code is ready and it has increased your, uh, velocity, uh, of the core changes in deployment. So you measure those aspects as well.

And what is the context switching with the ability to query code base and hundred knowledge in id? So basically the things which you are having the legacy core, and we don't have the any documentation around that. Those things gets resolved in no time.

So, um, what we have been told by our customers is that this enhances is accelerates 80% of the development task. Uh, there are 60% of the code acceptance rate. Yeah.

Uh, and when we say 60% rest, 40% is, hey, you know, but I have a very specific business requirement, which I need to do, which I need to code. Um, so what we are saying is, but most, and most developers are saying, Hey, you know, 60% of the core as it is, is coming, which is a great, um, uh, and that's something which has been coming out from, uh, one of our customers, which we were going through as well at the end. Um, so it's a real, it's a real numbers from a customer, and the developers are quite happy because they are now developing new features, um, a 40% productivity increase, which has happened around they, the task which they were not liking, like writing unit editing or, uh, doing the, uh, comments, uh, documentation that gets resolved automatically and security fixes and upgrade, uh, all those aspects so they can focus on developing the new features, which was where the developers trends were and what business was looking for.

As I was saying earlier, that GitLab deal, this has been integrated with Amazon queue, so it has got quite advanced, uh, capabilities, which is in the DevX ops workflows, uh, which the developers use every day. So have a look into it. Uh, if you're using GitLab, that's something which is, uh, uh, great, great, great, uh, thing, which has been, uh, we have been doing behind the scene working with the key lab.

Uh, and it'll help you to innovate, um, and deliver quicker, uh, at every step of your journey. Um, so friction through the s DLCs and those aspects, which was there, it has been making now seamless. Uh, it's an AI powered experience, um, as we were talking about.

So it's available in preview very soon to be ga, uh, but feel free to start using it around on those aspects. We did the build, um, and then we were doing the operations, and then every time the operation there was an issue. Uh, it was again, going to the developers.

So, uh, I don't know if you have seen recently, but that is, um, in the CloudWatch. We have got AI ops now, uh, AI for ops, which helps you to tell saying, Hey, you know, um, you have obser, you have, you have observed an error into the logs, and based on that error on the logs, you can just create, um, uh, the ai, uh, uh, enterprise issues, and then you can track through your Jira and so on, which is all in build to the integration, and then it'll do its magic and find it out. What could be the reason this issue has happened?

What was the change which has gone in? Uh, is that the change which is stopping your system to perform well? Or what, what is the kind of, uh, things which has been, which needs to be corrected?

Um, so it's a great thing, uh, which we have been working on. Again, it's been powered by Amazon queue, so, um, and it's built upon the best practices and our support, uh, uh, mechanisms, which has been there. Um, man is and optimize.

So customer were asking, saying, Hey, great. Um, a lot of time around billing, how can I do the usage trends and intuitive visualizations? All these are now supported with the, um, AWS using the qs, and as I was saying, diagnose and troubleshoot error as well.

So you can go to the CloudWatch, use the IW AI ops and using the AIOps, you can see that, hey, uh, what could be the diag diagnose and what is the troubleshooting error which has been there and remediate quickly, uh, uh, those aspects. So what, what are the kind, so the thing was it, the tools are available, but then the other aspect is asking the tool what your problem is. So here are some of the queries which you can use.

So you can ask saying, Hey, Lambda, hey Amazon queue, what is wrong with my Lambda function? Um, and it'll try to go through every aspect of the Lambda function logs and so on, and it can figure it out as to what does it looks like, uh, customer asking, saying, Hey, you know, I at the click of my button in the non-techy language, uh, my program has been asking me saying, what is the forecasted cost for the rest of the year? Uh, we don't need to go through the number fudging and all those things.

We can just derive those aspects, which has been available onto the billing console, and we can provide that the cost of the breakdown. Uh, how much am I spending on the data transfer? Uh, show me alarms for my S3 buckets.

Uh, how can I leverage agents? How can I leverage my a PA gateway, which has been seeing errors to resolve those issues in no time? So all of those things can be helping with Amazon queue to operate the things as well.

The most important aspect that, uh, we have recently announced at the reinvent, uh, which was around the transformation of the workload. Um, and fundamentally when we talk about the transform, it's a slow modernization. There is a complex legacy systems, which requires a lot of, uh, uh, workforce in order to carry out any transformation.

There's a limited scalability options, which has been available, and hence it takes quite a lot of months and months in order to do any transformation. So we have specifically created some solutions which has been tailored for, uh, specific problems. Um, and as you can see that we have done now, uh, if you want to move from Java version, legacy version 7, 8, 9 to Java 17, you can start leveraging the Amazon Queue developer.

If you have been using dotnet framework and you want to move to dotnet core, you can do that so that your application has been portable from the legacy version of the net into the conet core, and hence, you can run it onto the Linux as well, giving you the business benefit back, uh, in terms of, uh, running the net applications onto the Linux platform as well, saving you the license cost of the windows, and also taking the benefit of ization and scaling and all those aspects, which comes around mainframe, which was the biggest, biggest, uh, challenge you can. Now, I'm not saying all the problems of the mainframe is going to be solved, uh, but what we have started getting was that from point where the biggest question was saying, Hey, I don't know what has been running on my mainframe. We can easily now generate the documentation for the mainframe applications, what has been running onto those mainframes, which then helps us to create the business, um, uh, move forward onto the HS technologies.

Uh, as you can see, last year, Broadcom came out and say, Hey, you know, VMware licenses are going to be changed. Uh, and there was a lot of things which was changing around the VMware. Uh, one of the pathway was migrating those VMware VMs onto the EC2.

So using the Amazon queue developers transformation capabilities, we are supporting now to migrate those workloads in no time from the on-premise VMware or VMC onto the, uh, EC2 instances. So that's something which is a great, great significant benefit, uh, for the transformation which our customers are facing. It can be into the id, it can be into the, um, uh, web browser.

What we are saying is that the customer is saying that we are four x times faster. We are saving 40%, uh, savings when we are doing this kind of a transformation, which is a significant number. If you look around from months and months, which we are talking about, it has reduced down, got everything, uh, been recommended to the customers, customer can select saying, Hey, yes, this is moving great.

I want to trans, I want to migrate thousand applications. And this thousand applications, um, uh, needs to be, uh, done. Previously, uh, it used to take months.

Now, in the recent, um, uh, migration of the thousand Java applications where we needed to move from, uh, Java eight nine to Java 17, we were able to do it in two days on average, 10 minutes basically. So the developer can put it and then they can start developing the new feature. Once all the development, the migration has happened, it comes back and shows to the developer saying, Hey, you know, your application has been upgraded.

Do you want to adapt all the changes? You say yes, and it'll take over from there and then progress the next steps of the function to integrate and deploy and deliver as well. Um, it saves us 4,500 years of development work and 260 million.

So this is based upon the real, real work which has been done at Amazon. It's a significant numbers, uh, if you look around, uh, the last leverage data in ai. So data is the core of everything, what we have been doing now.

So you can start writing in your natural language, the pipelines, you can build it onto the SQL queries, uh, actionable insights. So basically if you go to the QuickSight and you say, Hey, what is my, uh, projection for this month, which is looking like based upon this particular product which we have been using on two, you write in the natural language and it'll create you the, uh, output what you're looking for. You can analyze the data and integrate that data into the, uh, features, what you want to have it, uh, as a business.

And you can train to do the business, uh, ML model as well to train the ML models. Uh, Amazon sales maker has been widely used to develop your own models. Uh, you can do the development into the Visual Studio Code or any id, which you on to, and it's, it has got the step by step guidance, uh, in the no code truthing.

And hence you can leverage the data and AI together, uh, with, with zero expertise, um, and a full blown product to be available for, uh, building up any product based, uh, product database product. So, um, as I was saying, couple of business, uh, uh, recommendation or customers who have used it, uh, national Australia Bank, they're one of the largest financial institution in Australia and they have been using it. And what they are saying is that 50% of the code suggestions what has been made by Amazon Queue developers, our developers are just accepting it.

And, uh, this has significantly enhanced our productivity, delivering better service to our customers. Similarly, um, NOVA Comp, we are into the IT services, uh, as I was saying that, uh, you can do the transformation. So there was Java eight to Java 17 transformation and 10,000 lines of code.

Um, we have six 60% decrease in average in our tech depth. So it does all the thing for you. Uh, eight synchronically, take your code, go to put it into the three, uh, do the transformation of the project from Java to Java 17, along with the testing and so on.

One it is, once it is a compilable state, it tells you saying, Hey, you know, you want to see the code changes, which has gone through and it shows you significantly improvement in terms of the developer's experience, uh, and acceptance there. Similarly, Toyota has been using it as well, and as I was talking about, they had the Copa uh, mainframe and it's been used for the persona driven insights in order to develop the lot of documentation so that they can take the documentation and start doing the better things, uh, with respect to those capabilities. Volkswagen as well, um, as I was talking about the GitLab, uh, they are using GitLab and they're using the AI capability with Amazon Q Developer, which has been integrated, and again, a great feedback, uh, which has been given by the Amazon feedback.

Similarly, masteries, they have been using it, uh, they SecOps, uh, for the purpose of DevSecOps, and that's have been something which is significantly, uh, their developers are loving it. Uh, and common task has been left to the Amazon queue developer. So, uh, what you can do, so, uh, there are two versions of Amazon Queue developer, which is available.

Um, uh, one is you can use the free trial, uh, uh, and that is something which you can just register at Amazon and just like how you were using the free car of the AWS, you can start using this, uh, uh, as an AWS player. Uh, the another option is the Amazon Pro. So Amazon Pro is where we don't learn anything from your data.

This model has been tuned, fine tuned, um, and they will put the recommendations and so on. So, uh, the pro subscription is something which you require in order to be using the Amazon Queue developer, which is what our most enterprise customers have been using it, and plan your POC next time you're doing the upgrade. Don't just do the upgrade next time you're doing the security fixes.

Don't do any security fixes without using the Amazon True Developer. This will significantly improve and you can showcase to your, um, uh, leadership that how this Amazon True Developer is helping in order to make the things faster and better. Uh, and we are happy to support that in case anything is needed from the AWS.

With that, thanks a lot today. Uh, absolutely pleasure to talk with. Um, feel free to put any questions into the chat and I will address it.

Thank you. Thanks a lot.