 You are up next. Hey, thanks. Let me know if I can see my screen and the screen. I will just say it. Yes, I can. Please go ahead. You have four and a half, five minutes. Thank you. So what I'm going to talk about is how do you, how do you build a code search engine, which is primarily for small teams and, you know, for your own personal needs. At Grammar, we use GitLab to actually save all of our repository, to save all of our code to a repository, and then, you know, maintain CICD and push the code regularly and so on. But whenever someone wants to search for a code, right, what developers usually do is go to, go online, maybe search for a specific module and see, and then modify the search results to see. It's not exactly fitting their needs, and they go to their own previous code, previous code basis, and try to find the right sample and then, and then confirm it further. What I'm now proposing is a very fine-tuned search engine, which is specific to your need, which is what most of our use cases are. And show you how to do that using simple NLP techniques. So I'm just going to show you, if you want to use, if you want to define cache for my endpoints, and whenever you create your APIs, you don't want to fetch your data each and every, each and every time. You want to cache it for a certain period of time, if they are static files like JavaScript, or maybe some data files, you may want to cache it for a long time, maybe one year. So, and if you go to GitLab and search for, let's say, how do you cache? Right, because that's something you do. You realize that there are no reasons, and then you actually need to select some project. Let's say I will select some project called GraphStarts. Even here, it is going to show me a bunch of results out of which I have to find out what my YAML file is and then figure out the right snippet for that. So instead of that, what Discover provides is, you can just search for the keyword cache. And depending on whatever your index, it's going to prompt you results for those, from those specific projects. And you can click this specific link and look for the cache, and then just copy paste the snippet either from here or maybe from wherever you had the search engines. So how does this work? You can just go to the configure window and then add some specific project and then submit a request for indexing. This is the UI aspect of it. But in the background, this is what is happening. So let's say I've shown you the example for YAML files, but if you were to search for Python files, how do you do it? You take every Python file and identify all the functions in the specific script. And within each function, you then identify the function name, doc string, function calls within the specific function and also the method calls. You then stringify each of it, maybe split it and then identify if the function name is, let's say function underscore name, maybe split it by underscore, I mean all the snake case strings. And if you have camel case strings also, let's say function name with n caps, you also split that. You do the same for each and every string in the function name, doc string, and also the function calls and the method calls. And then you form the whole entire string, the whole document, right? Then you do a TFI idea on top of that. What that will then give you is a matrix of rows and columns. Each column will be your single word. Each row is a function you're trying to identify. Depending on how deep you want to go, maybe you want to just identify the file first and then you want to identify the specific function I'm looking for. So I hope you liked it. If you want to check out, go to bit.com. Thank you.