 Hello, everyone. Welcome to this talk. My name is Florian Huynikke. I work for Gina AI. It's an open-source company for new research, and I'm very glad you invited me. Thanks a lot. Very appreciate it. And today I would like to talk to you about one of our new products called Gina Now. And it makes it very simple for you to deploy your own search application within minutes. And normally people from big companies would be able to deploy such a service, but since we're working on this on their open-source domain, everyone can deploy their service just within minutes, and I'll show you in a second how this works. Here you can see a deployment of Gina Now. And it's using a birds data set, which allows you, in this case, to search for text and also for image. For instance, we can search for the cardinal, one of my beloved birds, and also you can search for images. And let's search for this colorful bird. And as you can see, we get similar-looking birds as output. Sometimes ago I took this image of a sparrow at Berlin Alexanderplatz. After uploading that image, I get similar sparrows as search result. Our open-source product Gina Now is, of course, not just for birds, so it can work with any general image data. In the following example, I show you how it works with fashion data. Here I deploy Gina Now using a fashion data set. And summer is coming, so I want to have some shorts. So let's search for black and white shorts for men. And these are the results. And actually, I like this one a lot. So this front-end is very good for getting you started right now, to immediately run some queries and to explore your data. If you're ready for the next step, then you can also integrate Gina Now into your product by using our REST API. Here you can see the open API specification, and this is highly usable for the front-end engineers during the actual implementation. So I hope I made you a bit curious on how you can use your custom data to deploy your personal new research case using our open-source product Gina Now. So let's directly jump into the terminal. The first thing we have to do is to install Gina. So we do pip install Gina. After Gina is installed, we can just type Gina to see the help text. And as you can see, there is a Now keyword to get your search case up and running. So let's type Gina Now start. Now Gina finds out that Now is not installed already, so it kindly asks you to install it. So let's copy this command, type it here. After the installation is finished, you can just start Gina Now. You can use the arrow keys to select one of our predefined data sets, but you can also choose to take a custom data set. In this example, we will just go for artworks. Here you can select the quality of your search case, and you can choose between medium, good, and excellent, but also have a mind that if you choose a better quality, then we have to deploy a larger machine learning model, which then takes more resource from your machine. In this case, I choose medium quality. Now we get asked to choose the deployment option, and since all of the clusters are cleaned up on this machine, we can just create a new one. Here we have to decide where to create it. There are some possibilities, for instance, to deploy it on Google Cloud, but in this example, we choose to deploy it locally. Setting up the cluster, and then also indexing all data, will take a few minutes. In this video, we don't have much time, so I do a fast forward. And here we are. We can see that Gina Now finished the deployment and responds with a URL to the search case. In this case, it's a local deployment, so the URL is also localhost. So let's click on this and go to the search case. After opening the link in the browser, you can see the Gina Now front-end, but this time with hard data indexed instead of birds or fashion. So let's just run the search query to confirm it's working. So it looks good to me. By now, using your own images will be very easy for you. You just have to type Gina Now start then the data attribute, and provide a path to your local image folder. That's it. I hope you liked the video. And thanks again for having me. And the open source AI community is just amazing. I hope to see you soon. So, bye-bye.