 I'm a little bit of the black sheep of speakers tonight. I have seen some brilliant technical presentations tonight. I will focus a little bit more on the business side of things, but I'm pretty sure you will be able to relate to it. My name is Felix Schüller. I'm in the data analytics space since the last 10 years. Today, I'm speaking to you as representative of the Bunda Alliance, you and Jeremy, so the Founders Alliance Black Area in Germany. So the Bunda Alliance is actually initiated by 70 big and medium-sized corporates in Germany that want to promote their region more for startups and innovation. Those 70 companies have very well understood that startups are actually a capitalist for data development. And today, I will talk about their newest program, because we are currently launching our data program. So what we want to achieve by it, we want to better bring together corporate partners and startups. So the corporate partners are more focused on the focus of the specific area in Germany than I'm talking about. But we are looking for startups who belong to global. And that is also the reason why I'm here, and why my colleagues in India and North America and all kinds of countries in Europe. So how it works in detail, and what there is going to be for you as a startup, what you tell me later, let us maybe start with why we actually want to create the data out. So when we were talking to data innovation startups or data science startups, all of them facing the same challenges. So they have to make sure their product matches the market. It's very nice to build a technology demo on what you can do, but you actually have to set a product. So it's important that this really matches the market that somebody is willing to pay money for. Second, if you want to train your models, you need high-quality data. And you need high-quality data, not of the last two weeks. You need quality data for a couple of years to make it work. So where is this data coming from when your company is just like two months old? And third thing, money, right? Cash flow, especially if you're dealing with a business to business area, it can take six months, 12 months until you find your first client who is willing to pay for your MVP. This is definitely a challenge to startups, especially in the data science space. But there's another side to the coin, right? Not only startups struggle in the data innovation space, but companies struggle. They also face challenges. So for example, they address the problems easily from a very technical point of view. So they go to one of the big vendors and look for the data science software that they have on shelf. And they buy it and then introduce it to the company. And then they think there's really data innovation. Of course, this doesn't work, right? So second, many companies are also not having a good data strategy. Where do they want to go from here, from today, in the future of my data in two years and maybe five years? And third, and this actually is probably my favorite, what's the return on the best of my data science team? I've seen this a couple of times in corporate. They hire three data scientists, put them into a room, tell them to do data science. And after six months, they complain that the return on the best of the students will be back. But at the same time, there's no clear vision where this company actually wants to go with the data scientists. And at the same time, the data scientists have no authority to change processes. Data flows, for example, in the corporation. So they get the data, they actually need to run their analytics. So here, the question is not so much on the skills of the data scientists. They are usually pretty good. There's more depressions like, where do they want to go with my team? So when we are looking at those three challenges of cooperating, we are looking at the three challenges of startups, we also have to feel that there's some kind of match here. We have very skilled, motivated people, startups at the same time have cooperated with a lot of data history. Why should we not bring them better together, right? And our idea is we want to build the data, the data to collect unique data assets, connect startups and corporate partners together with experts, and also enable and encourage partnership between those players in the long term, not only for once in a project, but really for a long term, for a long term partnership. So why do we want to go to the world region in Germany, right? World region in Germany is like, in the rest part, it's not so much on the international startup of Mac as, for example, Munich, but it has the longest history of industrialization in Germany, right? And we see there is huge potential in this area. It's highly populated. You find 25 million people there. This is basically Berlin and it's a newly combined. So what you see here, especially an opportunity for is industry data. So it's, this area is famous for mining, tourism, street production, mining, this already came to an end over the last decade, ever since. But still, the cooperates, many cooperates are around and they're sitting on the history of data of easily 20, 30 years sometimes. This industry data is mostly untapped from the data science perspective. So we see a huge potential if the startup is still in development. At the same time, because of the history of coal mining and street production, companies are very closely connected to each other. So your chances, if you're getting one client for your product in this network, the shops are pretty good that you can sell it to some unknown companies there as well. So just to name a few, maybe those names will not tell you anything, this is pretty fine for there. Kind of probably not too small in Germany. I guess very famous ones is where the vocal was already each one of the main developers. Yeah, the design needs a bit of an easy reference. Okay, not very interesting time. What is earlier for you with your style? We just launched our first open core with a bunch of data challenges you can apply for SSI. So what is happening on our webpage, you find real-world business problems, business potentials for our company partners. They come together with new data sets you can apply to those challenges. And of course, if this partnership works well and everybody's happy, you have very good chances to create a new client in the time of maybe a new investor for you. And, oh yeah, something that's also budget-connected to every single challenge usually around the next five years. So you find all those challenges on our webpage, data.tour, currently you find our first open core seven challenges that are really focused on data science. Please have a look. I will now introduce you to three of those challenges to give you a impression of what it's all about. So, relatively easy to understand. I think it's so easy to solve this one from one of our partners in the rural region that's still in the water business, actually. So what they have is they have a couple of thousand sensors in the lakes and canals around the rural area collecting information about water level, water speed, so they control, they can accordingly control their water pumps later in the process. This is really critical for them to understand like and also predict how water level will develop over the last couple of hours. So for their own prediction models, they need high quality data. But those sensors are of course outside, so they get dirty, they break, sometimes there's a connection problem or even some benefits that happen. So what they're actually having is a team of specialists that mainly focus on fixing the sensor data on the database and identifying problems saying, okay, somebody should drive out into the field checking this and that sensor because the numbers look kind of strange. But it's fixing this data for their own models is a little bit more challenging than just taking the average over the start and point that it seems to look right. There's a lot of experience connected to this. So this could be very interesting data science case to replicate this experience, this knowledge that those experts have with the data, with the model. Another case that's actually my favorite, it's about ground depression. So as I mentioned before, the Rua area has a long history of mine. So they get done 1,200 meters or something like this and it has a lot of mine that was happening in this area. What it means if there's maybe a structural problem in the ground because of this mining and this can generate problems maybe 20 years later, 15 years later, causing actually holes collapsing on the surface. This is really an issue and can be the data of the future. So what's considered an indicator of the structural problems in the ground that can affect the surface are actually small ground depressions. Can be two meters wide, 80, 70 meters deep. But they are hard to find, right? We have data, our partner has data, R2D data as well as like 3D data of this area. But finding those depressions is currently a manual work and this is very inefficient and we are looking for somebody who comes up with an algorithm to improve the situation to automate this. What you can automate, what is very well automated is already, if you for example, have a new small ground depression, right? You fly over, you fly off cameras over your area and from one way to another you have a small depression. This can be identified really because this is not a challenge. The challenge is, how can I do it when I have no history of this, like flying over the area for the very first time. It personally reminds me a little bit maybe of finding cancer in X-ray pictures, but it seems to be a little bit similar, but this afternoon we will hold it with the rest of the session. Another case, natural gas price prediction. There's a European exchange basically of gas. It's not so much, this case not so much meant to predict the gas price in the long run, more on the daily basis, like even intraday. So what does influence the gas price? Okay, that's the obvious one, right? Well, okay, please. Electric power price seems to influence the gas price actually in the past it was oil, but this is not the case anymore, so electric power seems to be a better indicator. Also, our two emissions. If you want to burn gas, you have to buy CO2 emission certificates. To neutralize, that's also their prices influence the gas price. And also when we have also data on plant maintenance and the gas needs of Norway and basically get candidate the information of, okay, in two weeks we'll have the method of maintenance so for a couple of hours and we know the gas going through the pipes, what means this might influence the gas codes. This is just a couple of ideas we have came up with, or factors that might influence this price, but you want it to also be free to use other ones that you can come up with. So there's three of the examples, three examples of the challenges that we have on our web page. Let me tell you a bit about the process actually here. We, I'm not an open data platform. We don't want to be a web page where everybody can upload this data and without any idea what's behind it or what can be useful for and just waiting for anybody to pick it up and do something with it from our experience so far as that those platforms are not working so well. What we want to do is that our partners provide the data assets as well as bring a specific use case with this data actually and those seven use cases right now in front of the web page. Where we also separate a little bit is from running data changes in the way of, for example, category doesn't. I really like those category changes. I participate with myself and a couple of them. I really love them, but there's usually more that really takes it all, right? A lot of people, a lot of startups have a lot of time into this and basically you get something out of this if you make it under the top three days. So in our process, selection will take place before you start working on it. So you can use all your resources, all your focus on this specific problem because you're the only one working on this. So it takes me also to the timeline. Our application phase was one in September and it's also still running in October. Our deadline is so 21st of October. Then we'll be about like 10 days selection process of the specific startups with the partner that's actually providing us data. And then it will work for around three months depending on the case, complexity of the case, on the specific use case, and then we will think about long-term cooperation depending on the outcome of this. We should apply which startups are we looking for? Again, we're looking for startups globally. You should be a startup in the data science field that would be for sure beneficial. One hard fact is we want to have you a legal entity for over six months. So it means like if you're a couple of students have an idea, this may be not the right program for you, maybe you need to do something else. So we want to have a legal established company. And of course, when you sign up to one of those challenges, please send us your application, contain your technical approach that you need to try on your own on the program, as well as your ideas, references from projects that may be already done. This is very beneficial if you want to be selected for a specific use case. Again, that line is 31st of October. Maybe you look at our web page at those seven challenges and say, okay, I have any data science field and I have some experience there, but maybe there's no use case right now that it's been filling me. And then please at least sign up to our newsletter because as I mentioned before, this is our first open call with the first seven challenges. The next open call we are already planning is probably beginning of next year, maybe end of this year. We will write the next open call with new challenges, new partners. So the probability is high that maybe then there's a challenge for you as well. So if you want to be notified, please sign up to our newsletter. Or if you know another startup, maybe that you are aware of that we don't know, that could be interested in one of our challenges, please notify them, let them know. Maybe this is a good match. So again, this is our webpage, data app.org, thanks for your attention. And I hope I will see a lot of sign-ups and applications in the future. Thank you. Thank you very much. Thank you very much. Thank you. Okay, I think Paul has a question for you. Just a quick one, I have two. The first one is, I love to see the slides on the previous one, you access a data with this tree. So I assume you have some sort of versioning with the data on your platform. Maybe you can share more on that. And then let's say if someone wants to interact with their data, let's say someone you have to send already in the process itself. When they interact and get access to the data, please sit by it, like if you provide APIs, or this is something that they download by a CSB, or this is something that they have a kernel running on your platform that you can access, especially for p-data, they cannot just download that or something. Wow, how old are those? Big question depends. There's other stuff in there. So what we wanted to achieve in the end is we want to build also bigger assets, right? So right now, those use cases come with specific data assets. You can request on the web page a data sample. We will send you some couple of rows, basically, of the data. The first look at, we will receive the full documentation about the data. You will not receive the full data set. The full data set will be available to the startup that will run on the specific product. But what we want to do is, of course, collect those data assets over time on our data hub. What means, if you maybe are not chosen for a specific problem, but you would say, OK, with this data, I could solve a completely different problem that is not part of the use case. Then please reach out to us. We will manage it to get access to the data. This is the idea. We want to connect basically people. We want to connect also startups also to the right persons within the corporate. Because from our experiences, it can be quite hard. Like, if you don't know anybody in the corporate world and it's a corporate data firm, it's really hard to open the door, to gain trust, to talk to the right person. We want to take this up. We want to be there as a mediator for you. Like, connecting you to the innovation officer, whoever, who's in charge of the data, who can basically decide which data is useful for you, and there's even additional data that we haven't thought about. So we really want to be there as a mediator between both sides. Data, it depends, right? So we receive a lot of data from a CC format, extra format as well, if they are not so big. We have some 3D-point data, GIS data. Format-wise, it really depends on the use case and the kind of data. You will always receive their documentation files that basically explain to you what it is. We really want to have data in standard formats, so not something super fancy where you have to buy a specific software for $20,000 if it's not that big. Any more questions? Anyone has questions for the next? OK. If not, I think we can end our today's session. Thank you very much, please.