 Okay, great. So virality became a holy grail of the internet that everybody's chasing around. Publishers, politicians, marketers. And I'm a publisher myself, and I was thinking hard, how can I expand my audience? And the truth is, in order to get any visibility on the internet, your content should be extremely trendy and readable. And I was trying to look for data-driven solutions to reinforce my content strategy, but I didn't find anything useful. That's why we decided to create such a tool by our own, where you can upload your text and get your virality score. We've been studying more than 10,000 factors that influence the popularity of the text. We found the combination of 200, the most impactful factors. And just a half a year ago, the accuracy of our predictive models was 55%. Right now, it's getting closer to 90%. The next goal is to create an AI co-writer that will suggest you real-time how to improve the publications, and which will increase the likelihood of going viral. The value of such tool is pretty straightforward, because one additional thousand of views brings one additional, at least five additional dollars of a bit of a thousand revenue. Of course, there are plenty of content optimization tools on the market right now, but none of them is dealing with pre-publication stage. Only large publishers like BuzzFeed and Mashable are dealing with the similar technology, but we aim to democratize it in the era of attention economy. Michael Fowder is a lead data scientist at one of the biggest gaming companies in the world, Wargaming. He holds several scientific patents, and we're looking for angel investments and early adopters to experiment together with our technology. Thank you very much. That's it. Thank you. Do we have any questions, comments? Sure. In terms of cases, would it be called to read a text you've created? Could you explain one case that you have done that was successful in terms of... Yeah. Well, right now we are experimenting with our first customers. We're creating data, we integrate API on a backland level, and yeah, they're basically using our predictive score, and right now we are about to measure how much value it doesn't bring. So we are in the process. I mean, this is like clickbait to the next level in a way. Well, not really, because what we figured out, there is no universal notion of virality itself. So we optimize the content for the target audiences. So if you write for TechCrunch, we make it viral for TechCrunch audience, not for BuzzFeed. So you're connected to the actual publisher and do some language analysis. These are the words that work, kind of. Yeah, we analyze audience interests, and then according to this interest, we adjust virality. Okay, cool. Thank you. This is a separate solution for a plugin for existing publishing systems. Existing solutions? No, as we know, there is no existing solutions. We know that BuzzFeed has pound and Mashable has velocity. They're doing for their own content. But we are definitely among the first who is dealing with this. But this is not integrated into the existing publishing systems. So it's a separate service? Yes, it is. Okay. What are your target customers and what's your go-to-market strategy? Right now, we're working with publishers as early adopters, but our target market is native advertising market, content marketing mainly. So how are you going to get the first customers? Well, we already have lots of customers in the waiting list. We just don't have capabilities to deal with them. So right now, we just choose several customers. But yeah, as soon as we deliver the final product, we will be able to deliver it. Hi, is it possible to just give one score? Or do you like measure different kind of... I mean, you said yourself, you measure them in 10,000 different categories and issues, right? Can you elaborate a bit more on what those 10,000 would be? Because I felt a bit understand how it does it. I mean, it can be too complicated to explain in two minutes, but a bit more dig into that. Well, we've been studying factors like sentiment analysis, like tone of the article, trends that we get from Twitter and Facebook, all kind of grammar aspects and so on and so forth. So these are the main categories. And you're based in Belarus or you're based... Currently, I live in Madrid. But the tech team is in Minsk, for sure. Cool. And what languages do you support now? I guess English, do you support other languages as well? Yeah, right now it supports only English language. That makes sense. I think this sounds super interesting, but also super scary at the same time. But very interesting. Thank you. More questions? All right, if we don't have any more questions, let's give a big hand for Daria from Expanenta.