 Hey, everyone, and welcome to my webinar on how we can use AI to reduce friction in the DevOps day-to-day activity. My name is Eran Bibi. I'm the co-founder and Chief Product Officer at Firefly. In Firefly, we are helping teams to control their cloud by automatically convert their cloud into IAC. Before that, I was the head of DevOps and SAS platform and Aqua Security, and I'm doing DevOps for over 10 years. I really like to learn new technologies such as AI and then explain them to others. In three years' time, AI will be taking over the world. Just before that, let's try to have some fun with it. So what is the first thing come to your mind where we are asking what AI is capable of? You got it right. AI can create amazing images. And this is exactly what we did at Firefly. We ask AI how our beloved tools in the DevOps ecosystem will look like if they were human. So here we have Jenkins, Crossplane, GitLab, and Argo CDs look as human. The results are pretty amazing. If you like it, we have 20 more and we publish the results in our blog so you can just scan the QR code and see all of the other tools. So what else? It can generate images, but it can also write marketing emails. It can write blogs. It can plan our next vacation and it even can create the ingredients of our beloved dish that I would like to cook this evening. But what make it more fascinating is that AI can generate code. And if we are taking this amazing capability and incorporated with the DevOps day-to-day work, we can find some very nice use cases that we can leverage AI for our behalf. So AI can generate IAC templates. It can do terraformed cloud formation and even imperative languages like HuluMe. It can also create configuration files like Docker also Kubernetes manifest. If I'm using my pipeline that is described as code, it can generate pipeline for you. For example, Jenkins files and GitHub action. It can also generate policies, shell scripts, Python utilities. It can build your command line. If for example, I have some complex COOP CTL command that I would like to have. But don't need to use any cheat sheet or documentation. I can just ask it from JetGBT. I will get the command line built for me. I can also create a complex queries to get data from my elastic, Mongo, MySQL or any other database. So let's see some example. I went to JetGBT and ask for Docker file for my Node.js. I also mentioned that I would like to have it secured. So I see the baseline image is based on Alpine, which is great, it's secure and minimal. Another example, I ask for a MongoDB deployment getting a Kubernetes manifest for it. So if I'm getting a response that is not identical to what I'm looking for, I can use the chat to reiterate. So I can ask for a specific thing. For example, if I would like a different number of replicas, I can just ask it and get a different result and minimize the scope to get an accurate response from JetGBT. I can also ask for policy, if I'm using policy as code framework like OPA, I can ask for policy to enforce that I have a readiness probe on all of my Kubernetes deployment. And of course I can take this policy snapshot and put it in a gatekeeper or other system that will make sure everything that I'm putting in my cluster will be according to my policies. So JetGBT is amazing tool. It can generate code, manifest, scripts, whatever you are asking for. But then you need to copy it and paste it to make sure it's working. And it's not that convenient work with browser-based solution. And this is exactly why we introduced open source project called AIIC. So AIIC is a CLI. You can install it locally in your Mac or remotely in one of your Linux VMs. And then you can basically prompt the AI to get responses back to your local machine and save it directly on your local disk. So this is very convenient and very easy to use. And you can also have an automation flow around it because everything is done with a command line. So let's go back to the use cases. You can ask AIIC to generate Terraform code. You can ask for it for configuration files like templates, CICD pipelines and all of the stuff that I already mentioned, everything directly from your command line, which is pretty awesome. So let's see exactly how it's look like. So in my terminal, I already have AIIC installed. Now I will ask AIIC to generate Jenkins pipeline for me. So I'm going to do like AIIC get Jenkins pipeline to build node back. It will take a few seconds for it to generate it. Let's see what are the results. So I have a very basic Groovy template here. There is few stages, the checkout, getting the source code, the build, test and deploy. And of course, if I'm not pleased from the results that I just get, I can put R, which is retry. Let's see if I'm getting different results. Great, now I get another type of pipeline, which is Jenkins declarative. You can see here, I have the environment, the stages are in a block of stages. Seems something, let's assume, are pretty satisfied from that result. What I can do is just put save, give the name of the file or the full path, Jenkins. Okay, I have Jenkins file here locally in my computer. I can see this is exactly what I got. And this is amazing. And of course, you can have AIC working with a non-interactive mode using the flags. So you can mention the path directly from the command line and it will automatically save it without interacting with the user. Let's see another example. AIC get cube CTL command to spin up and Genix up. So in this case, I would like, I don't remember the command syntax for getting a quick and dirty, let's say, an NGNX application running in my cluster. So I'm asking just chgpt to prompt it to me. And within a few seconds, I got the command line. Cube CTL create deployment, NGNX minus minus image, this is the flag to mention the exact image. In this case, it puts the latest image of the NGNX container, taken from Dockerab in this case. So these just few examples of how you can use AIC in your day-to-day. And of course, I'm going back to explain more about AIC and how to get it. So AIC.dev, this is the place to collaborate with AIC and of course, we are based on GitHub. So we have almost 2000 stars in the project and feel free to PR any kind of changes and comments. And we are the next version of AIC will use the new chat GPT API that was introduced last week. So we are looking for that release. We will add few more capabilities to have more chat based like interaction to AIC. So this is something coming very soon, the upcoming game, upcoming days. Feel free to use that and thank you. Hope it will be, this session was useful for you. Take care.