Latest / Deep Tech Germany – AI, Robotics & Frontier Innovation by Startuprad.io™ / QC Design's Ish Dhand: Making Quantum Computing Economical
Transcript
- WEBVTT 00:00:00.000 --> 00:00:19.440 Music. 00:00:20.033 --> 00:00:23.473 Hello and welcome everybody. This is Joe from StartupRate.io, 00:00:23.613 --> 00:00:29.073 your startup podcast, YouTube blog and internet radio station from Germany bringing 00:00:29.073 --> 00:00:32.113 you news and interviews from the German startup scene. 00:00:32.193 --> 00:00:39.853 Today, I'm delighted to kick off again our fourth cooperation with the German 00:00:39.853 --> 00:00:45.753 Startup Association where we interview the winners of the German Startup Awards. 00:00:46.593 --> 00:00:51.793 Today, I would like to welcome the Newcomer of the Year, German Startup Awards 00:00:51.793 --> 00:00:55.233 2024, Isch. Hello and congratulations. 00:00:56.113 --> 00:01:00.073 Thanks so much. It's my pleasure to be here. It's awesome that you are here. 00:01:00.113 --> 00:01:02.893 We already had a lot of fun before this interview started. 00:01:04.253 --> 00:01:07.893 Maybe we can give you a sneak peek into what we've been talking about. 00:01:08.113 --> 00:01:12.213 But first, as always, we'll be talking a little bit about you. 00:01:12.213 --> 00:01:17.673 You are joining me from lovely Ulm, but you have not always been there. 00:01:17.753 --> 00:01:20.433 When I've been looking in your LinkedIn profile, by the way, 00:01:20.453 --> 00:01:26.033 as always linked in our blog post, you studied in Canada, University of Calgary, 00:01:26.253 --> 00:01:31.493 India, India Institute of Technology in, how is it pronounced, Canpur? 00:01:31.913 --> 00:01:36.153 Canpur, yes. Canpur. And in Germany, University of Ulm. 00:01:36.193 --> 00:01:41.633 And as all Germans, we always have the tongue twister in mind in Ulm, Ulm, Ulm, and Ulm, Ulm. 00:01:42.213 --> 00:01:45.333 Can you take us a little bit through this journey, including, 00:01:45.473 --> 00:01:51.273 I've seen you have been awarded the National Child Award for Exceptional Achievement 00:01:51.273 --> 00:01:55.733 by the Prime Minister of India in 2006. Thank you. 00:01:56.258 --> 00:02:00.138 You must have done something outstanding. Can you tell us a little bit about 00:02:00.138 --> 00:02:04.438 this before we dig a little bit deeper in what you've been doing? 00:02:04.598 --> 00:02:07.418 Plus, did you have a lot of warm socks in Calgary? 00:02:09.298 --> 00:02:12.078 Let's start with the last question. Yes, warm socks. Absolutely. 00:02:12.718 --> 00:02:14.958 The best warm socks are what you get in Calgary. 00:02:15.498 --> 00:02:18.758 Yeah, talking a little bit about how it was like growing up in India. 00:02:18.838 --> 00:02:22.578 I think I'm really grateful for a lot of effort put in, for example, 00:02:22.678 --> 00:02:25.758 in finding young talent. 00:02:26.478 --> 00:02:30.058 And giving them opportunities that are otherwise hard to come across. 00:02:30.258 --> 00:02:35.358 So there were a couple of opportunities for me as in, let's say, 00:02:35.418 --> 00:02:41.698 ninth grade or tenth grade, where we had the possibility to be challenged in science creativity. 00:02:43.038 --> 00:02:48.038 So lots and lots of top scientists would come together with these young kids 00:02:48.038 --> 00:02:50.538 from the age of 10 to 20 or so. 00:02:50.618 --> 00:02:55.298 And then they would basically challenge these kids to do something creative, 00:02:55.298 --> 00:02:58.498 which means from everyday objects 00:02:58.498 --> 00:03:01.338 that you find around the house can you make something that's 00:03:01.338 --> 00:03:04.478 truly new not just reproducing textbook experiments 00:03:04.478 --> 00:03:07.858 but make something truly new and this was the national 00:03:07.858 --> 00:03:10.958 Palshri award and here the 00:03:10.958 --> 00:03:13.858 award was given by the president of India back then 00:03:13.858 --> 00:03:16.498 this was APJ Abdul Kalam who is one of 00:03:16.498 --> 00:03:19.478 the top rocket scientists in the world and spent a 00:03:19.478 --> 00:03:22.378 lot of time in NASA and then also set up the 00:03:22.378 --> 00:03:25.138 rocket program in India or really took it forward quite a bit 00:03:25.138 --> 00:03:27.938 and following from that there was the 00:03:27.938 --> 00:03:30.938 national child award for exceptional achievement by the prime minister of india 00:03:30.938 --> 00:03:35.038 where it was really a similar situation from every province in india they chose 00:03:35.038 --> 00:03:40.978 one kid for basically going above and beyond and exports expose them to opportunities 00:03:40.978 --> 00:03:44.738 which otherwise quite hard to get so those were really formative experiences 00:03:44.738 --> 00:03:47.358 for me growing up in India. 00:03:49.338 --> 00:03:53.198 I see. And how did you end up in Calgary? 00:03:53.358 --> 00:03:59.578 Because being from India or from Germany, you don't think about studying in Calgary. 00:03:59.718 --> 00:04:05.278 You have other universities, especially in the US and the UK or even in Australia in mind. 00:04:05.358 --> 00:04:08.618 How did you end up in Calgary and how did you like it? Interesting. 00:04:08.898 --> 00:04:11.998 So I did my undergrad, as you mentioned, in IIT Kanpur. 00:04:12.218 --> 00:04:18.458 This was an amazing place. I focused as an undergrad on doing computer science. That was my major. 00:04:18.578 --> 00:04:22.938 And it was basically one of the top programs in the world. And also, 00:04:23.018 --> 00:04:24.998 I was very interested in physics. 00:04:25.058 --> 00:04:30.298 And Kanpur gave me this possibility of being at the frontier of both computer science and physics. 00:04:30.538 --> 00:04:35.618 During my time there, I learned something very cool about quantum computing, 00:04:35.738 --> 00:04:41.978 which is that if any civilization anywhere in the world is building the ultimate 00:04:41.978 --> 00:04:43.058 computing machine, team. 00:04:44.015 --> 00:04:48.515 Then this computing machine is basically a universal quantum computer. 00:04:49.275 --> 00:04:54.255 So I was very interested in quantum computing during my time in my undergrad at IIT Kanpur. 00:04:54.755 --> 00:04:59.735 So starting out from computer science, then the next stop was to basically look 00:04:59.735 --> 00:05:05.375 for a physics master's or PhD and try to explore this new field of quantum computing. 00:05:05.515 --> 00:05:10.075 I applied, of course, to top universities in the US, lots of good universities 00:05:10.075 --> 00:05:14.015 in Canada. And basically, I got rejected from every single U.S. university. 00:05:14.315 --> 00:05:18.435 I think from their perspective, I really understand what is this computer science 00:05:18.435 --> 00:05:20.615 guy trying to do here in the physics department. 00:05:21.395 --> 00:05:29.575 But then in Calgary, there is a strong core of quantum computing researchers, 00:05:29.935 --> 00:05:32.775 including my Ph.D. supervisor, Barry Sanders. 00:05:33.015 --> 00:05:38.495 And Barry really saw the potential in working at the interface. 00:05:38.495 --> 00:05:43.955 And then Barry took me on as a master's student very quickly switched to a PhD 00:05:43.955 --> 00:05:49.095 program in three years or so finished the PhD did some really incredible work 00:05:49.095 --> 00:05:53.335 with Barry at that time I was really happy with what came out of the PhD and 00:05:53.335 --> 00:05:55.695 yeah that's that's how Calgary happened. 00:05:55.935 --> 00:06:03.335 I see so they're very good in quantum computing uh-huh we may add that the India 00:06:03.335 --> 00:06:10.975 Institute of Technology IIT is one of the premier institutions of educations in India. 00:06:11.215 --> 00:06:19.935 And now we have this piece. Then it doesn't take a much longer way to understand 00:06:19.935 --> 00:06:26.415 how you ended up in a quantum computing unicorn. 00:06:26.635 --> 00:06:30.875 But first, let us take a little break for our sponsor. our sponsor, 00:06:30.955 --> 00:06:37.015 StartupRaven.com, the best way to find investors and cooperation partners for early stage startups. 00:06:37.255 --> 00:06:42.375 You can sign up for early access at StartupRaven.com. And we may tease that 00:06:42.375 --> 00:06:45.515 we've covered the German Startup Awards since 2021. 00:06:46.175 --> 00:06:50.335 So in the fourth year now, and down here in the show notes, we will give you 00:06:50.335 --> 00:06:55.235 a link to a playlist of 21, 22, 23, and you're the first one of 24, 00:06:55.235 --> 00:06:57.415 but hopefully not the last one. 00:06:57.575 --> 00:07:07.295 So, as I was teasing, you ended up working in Toronto at the quantum computing unicorn Xanadu. 00:07:08.595 --> 00:07:13.095 Was this a connection from the university? Not so much from the University of Calgary. 00:07:13.255 --> 00:07:16.855 So, I think we moved to Xanadu. 00:07:18.305 --> 00:07:23.985 But maybe one can talk a bit about from Calgary, I actually then moved to Germany. 00:07:24.705 --> 00:07:31.325 And I think the German experience is actually then what led me to Xanadu. 00:07:31.465 --> 00:07:34.805 So talking a bit about that, that is an interesting twist. 00:07:35.365 --> 00:07:40.205 You came from India to Calgary to Germany, only to go back to Canada. 00:07:40.805 --> 00:07:44.365 Absolutely. Quantum computing is an international is a very, 00:07:44.405 --> 00:07:47.945 very international activity right now. We're all working at the frontier of human knowledge. 00:07:48.305 --> 00:07:53.645 So geography, I think, doesn't stop people from basically working on cool things. 00:07:53.965 --> 00:07:58.885 Okay, okay. So you came to Ulm here in Germany, right? Yes, absolutely. 00:07:59.065 --> 00:08:03.565 I spent around three and a half years doing a postdoc with Martin Plenio, 00:08:03.685 --> 00:08:08.325 who is now the co-founder of QC Design, actually, the startup along with me. 00:08:08.665 --> 00:08:13.205 And so in those three and a half years, I think one of the things that happened 00:08:13.205 --> 00:08:20.025 was we started to see quantum computing as a field go from something that happens 00:08:20.025 --> 00:08:24.565 only in the lab to something which is actually of true potential and something 00:08:24.565 --> 00:08:27.805 where actually humanity can start to use the fruits of quantum computing. 00:08:28.025 --> 00:08:30.525 This was the time from 2016 to 2019. 00:08:31.765 --> 00:08:36.825 I was fortunate enough to be able to work with Martin, more on that in a moment, 00:08:36.925 --> 00:08:41.585 but also to be able to work with the top experimental groups in the world and 00:08:41.585 --> 00:08:44.425 really see this technology coming out of the lab. 00:08:44.685 --> 00:08:47.425 So it was only natural that basically by the end of my time there, 00:08:47.565 --> 00:08:51.425 I wanted to do something useful, something that actually we can build, 00:08:51.545 --> 00:08:54.545 you know, a real quantum computer that's useful. 00:08:54.925 --> 00:08:57.285 And Xanadu was the best place for this. 00:08:58.285 --> 00:09:02.285 I see. And there you actually headed a team, right? 00:09:02.465 --> 00:09:05.825 Yes, I headed the architecture team, one of the three teams. 00:09:05.945 --> 00:09:09.585 Very proud of the work that the team did there. So one of the things that we 00:09:09.585 --> 00:09:16.185 worked on was to come up with the first true blueprint for building a useful 00:09:16.185 --> 00:09:17.505 quantum computer with light. 00:09:19.405 --> 00:09:24.265 And it was an amazing experience because in six to nine months, 00:09:24.365 --> 00:09:31.045 basically, we put together a team of top researchers in that specific field in whatever capacity. 00:09:31.045 --> 00:09:36.225 We were able to align this team towards working on one single goal, 00:09:36.385 --> 00:09:40.205 which is, I think, quite non-trivial because often we are working with top researchers 00:09:40.205 --> 00:09:41.525 who are very independent-minded. 00:09:41.865 --> 00:09:46.725 And with that aligned team, we were able to work together and deliver something 00:09:46.725 --> 00:09:53.905 which really pushes forward how humans think about quantum computing. 00:09:54.165 --> 00:09:58.425 So that blueprint project is something I'm very proud of in terms of how well 00:09:58.425 --> 00:09:59.625 the team accomplished it. 00:09:59.625 --> 00:10:07.265 Before we get into that so quantum computing is based on quantum theory on the 00:10:07.265 --> 00:10:13.765 works of Max Planck and the quants are little crazy pieces so small the smallest 00:10:13.765 --> 00:10:16.285 we can investigate right now and. 00:10:17.351 --> 00:10:22.211 I do believe you can only see those pieces. When you smash atoms, 00:10:22.351 --> 00:10:29.971 we have very extremely fine instruments, only then it is even possible to investigate them. 00:10:30.071 --> 00:10:36.531 And that is basically the level when the laws of physics that govern our world 00:10:36.531 --> 00:10:42.151 are actually breaking down, as I understand, they have a randomness to them. 00:10:42.251 --> 00:10:47.271 And that always reminds me of Heisenberg in Breaking Bad. And also, 00:10:47.331 --> 00:10:54.051 I do remember quants, they do have, some of them have spin and taste and something like that. 00:10:54.151 --> 00:11:00.051 Can you give us a very, very dumbed down introduction into quantum computing? 00:11:00.251 --> 00:11:05.351 And then after we've done that, we'll talk about quantum computing with light. Okay. 00:11:05.591 --> 00:11:10.551 Sounds good. Sounds good. Let's talk about quantum physics for a moment. 00:11:10.731 --> 00:11:15.811 So there are two aspects here. There is what I would call quantum physics 1.0. 00:11:16.311 --> 00:11:21.931 Where we start to look at the behavior of individual atoms or individual particles of light. 00:11:22.031 --> 00:11:27.871 And this kind of physics actually underlies a lot of the technology that we use today. 00:11:28.031 --> 00:11:33.991 So the laser in the supermarket or the silicon chip in the computer that you 00:11:33.991 --> 00:11:38.291 and I are using, these all use quantum effects already. 00:11:38.291 --> 00:11:41.351 But here we are just using the fact 00:11:41.351 --> 00:11:46.031 that quantum states can be in superposition 00:11:46.031 --> 00:11:51.591 which is just a big word for saying that it can be in the state 0 in the state 00:11:51.591 --> 00:12:02.551 1 at the same time unlike a classical bit which can either be 0 or 1 but not both at the same time. 00:12:03.171 --> 00:12:10.271 Sorry, one question. If you rely on those qubits, it sounds a little bit like 00:12:10.271 --> 00:12:11.771 it's a bit with Alzheimer. 00:12:12.331 --> 00:12:18.111 How can you work with that? This is actually a very deep question and interesting. 00:12:18.990 --> 00:12:23.930 Working with these fragile, almost random bits, random qubits, 00:12:24.030 --> 00:12:29.070 is one of the big challenges to building a large-scale, useful quantum computer. 00:12:29.450 --> 00:12:32.970 And the way we work with it is in spirit. 00:12:33.070 --> 00:12:38.270 It's very similar to how, for example, the Wi-Fi that we use, 00:12:38.330 --> 00:12:39.310 the Wi-Fi connections that we 00:12:39.310 --> 00:12:44.510 use, can still work despite there being some interference along the way. 00:12:44.730 --> 00:12:50.570 The answer is called error correction. So you use many, many quantum bits to 00:12:50.570 --> 00:12:57.510 act together as a single but very reliable logical qubit. 00:12:57.650 --> 00:13:02.230 And this logical qubit is basically what one would use for useful quantum computing. 00:13:02.970 --> 00:13:13.010 And still there are many quantum pieces there working together as a qubit. 00:13:13.010 --> 00:13:17.470 They're still faster and smaller than 00:13:17.470 --> 00:13:21.390 a normal bit would be yeah so a single 00:13:21.390 --> 00:13:27.750 logical qubit could comprise somewhere between 10 and 10 000 physical pieces 00:13:27.750 --> 00:13:35.030 physical qubits so there's a big overhead here and nevertheless there are some 00:13:35.030 --> 00:13:39.310 problems for example if you want to study a molecule or study 00:13:39.610 --> 00:13:44.930 a material or break encryption, for example, there are many problems where quantum 00:13:44.930 --> 00:13:48.490 computers can do things which classical computers just cannot do. 00:13:48.650 --> 00:13:53.370 And this is why there is great promise in a new way of computing, 00:13:53.550 --> 00:13:55.530 basically with logical qubits. 00:13:56.750 --> 00:14:00.210 And what, for example, would they be able to do? 00:14:02.420 --> 00:14:08.340 The first applications of quantum computing will be in simulations of materials 00:14:08.340 --> 00:14:10.380 and molecules. But what does that mean? 00:14:10.940 --> 00:14:18.380 If we want to study the properties of a drug on a computer before we ever start 00:14:18.380 --> 00:14:22.520 to make it in the lab, this is something a quantum computer can help with. 00:14:22.700 --> 00:14:26.840 If you want to come up with a new kind of a material for an electric vehicle 00:14:26.840 --> 00:14:32.040 battery, this is something which can be done in simulations on a quantum computer 00:14:32.040 --> 00:14:36.560 much, much faster than what a classical computer could do, even a supercomputer. 00:14:36.960 --> 00:14:42.000 So these are the first applications. But only when we actually start to build 00:14:42.000 --> 00:14:45.840 truly useful quantum computers will the next set of applications get unlocked. 00:14:46.260 --> 00:14:51.480 I see. So we're now a little bit in quantum computing. 00:14:51.480 --> 00:14:58.920 And now, can you tell us a little bit about the world of quantum computing and 00:14:58.920 --> 00:15:04.820 what your company, QC Design, that he co-founded, is actually doing? 00:15:04.900 --> 00:15:09.060 So we're now drilling down to the core of the matter here. Right. 00:15:09.600 --> 00:15:15.480 So as I mentioned, to make a useful quantum computer, you need to put many physical 00:15:15.480 --> 00:15:23.340 qubits together into a logical qubit. but this requires really sophisticated designs. 00:15:23.800 --> 00:15:28.860 There are an infinite number of ways of putting qubits together, 00:15:29.020 --> 00:15:36.840 of controlling them with, for example, microwave pulses or actually looking 00:15:36.840 --> 00:15:42.940 at the qubits and using that information to figure out where an error happened. 00:15:43.180 --> 00:15:46.220 So all of these designs together this is 00:15:46.220 --> 00:15:49.720 called an architecture and there are 00:15:49.720 --> 00:15:52.440 many hardware manufacturers in the 00:15:52.440 --> 00:15:55.380 world right now who are working on their first 00:15:55.380 --> 00:16:00.380 architectures for useful quantum computing on their first demonstrations of 00:16:00.380 --> 00:16:06.760 useful quantum computing this is the endeavor that we at QC design support so 00:16:06.760 --> 00:16:09.800 we provide very powerful design 00:16:09.800 --> 00:16:14.420 software you can think of this design software as for example AutoCAD, 00:16:14.520 --> 00:16:17.900 or if you come from the semiconductor industry, 00:16:18.160 --> 00:16:23.820 then you can think about design software similar to what is provided by Cadence and Synopsys and Ansys. 00:16:24.320 --> 00:16:29.240 But this design software gives our customers superpowers. 00:16:29.360 --> 00:16:33.560 So it's researchers and architects within quantum hardware manufacturers that 00:16:33.560 --> 00:16:40.440 use our software to come up with the most efficient ways of building I see. 00:16:40.620 --> 00:16:45.680 So to put it in very, very simple terms, you're kind of the computer assisted 00:16:45.680 --> 00:16:52.240 design that architectures would have for their houses, but you are this piece 00:16:52.240 --> 00:16:55.240 of software for quantum computing. Absolutely. 00:16:55.780 --> 00:16:59.240 I think now a lot of people understood here. And I'm going to ask you a question. 00:17:00.103 --> 00:17:05.883 We skipped over a few pieces here, but in the preparation, we had a lot of text 00:17:05.883 --> 00:17:08.503 here, and we went over some of them. 00:17:08.703 --> 00:17:13.263 But going a little bit back to qubit and quantum computing, 00:17:13.703 --> 00:17:23.743 according to your own website, you participated in a world record tomography of a 14 qubit computer. 00:17:24.363 --> 00:17:28.103 Did you end up in the Guinness Book World of Records with that one? 00:17:28.823 --> 00:17:32.363 Unfortunately not. It's not yet a category, but maybe one day it'll be. 00:17:32.843 --> 00:17:36.823 I have to say, this was really a part of a big team effort. And there were many, 00:17:36.883 --> 00:17:40.503 many people who were working much harder on this goal than I was. 00:17:41.463 --> 00:17:46.083 But it was a very interesting experiment, which we performed with collaborators 00:17:46.083 --> 00:17:51.083 in Innsbruck and along with basically my research group where I was working at Ulm. 00:17:51.923 --> 00:17:57.143 We had come up with a way to solve an interesting and important problem in quantum computing. 00:17:57.843 --> 00:18:01.823 So, as I mentioned, quantum computing can do things which classical computers 00:18:01.823 --> 00:18:03.303 just cannot do, which is great. 00:18:03.703 --> 00:18:07.983 But it also poses some challenges along with it. And the challenge is that it's 00:18:07.983 --> 00:18:12.823 really hard to figure out whether the quantum computer is doing the right thing. 00:18:14.023 --> 00:18:15.543 It sounds like AI. 00:18:16.843 --> 00:18:19.023 Yes, perhaps even worse. Who knows? 00:18:20.623 --> 00:18:25.223 Okay. And tomography here just means that given a quantum computer, 00:18:25.443 --> 00:18:29.323 can you do something to it to figure out what actually it is doing? 00:18:29.523 --> 00:18:33.943 And this was a very efficient algorithm developed in the research group where I was working. 00:18:34.103 --> 00:18:39.863 And we worked with one of the top research groups making these quantum computers 00:18:39.863 --> 00:18:42.683 to realize this tomography algorithm. 00:18:43.723 --> 00:18:48.223 And yeah, it was a very positive result that the quantum computer was doing 00:18:48.223 --> 00:18:53.323 more or less what we thought it should be doing. and this was on a 14 qubit 00:18:53.323 --> 00:18:54.323 quantum computer back then. 00:18:54.403 --> 00:18:58.863 I do think this was 2019 or so and then since then the field has moved on. 00:18:58.963 --> 00:19:04.623 We have way more qubits out there in the wild today and many more powerful quantum computers. 00:19:06.306 --> 00:19:12.526 Mm hmm. And going a little bit back on what you and QC design are doing. 00:19:12.626 --> 00:19:17.046 How did you actually start this idea? 00:19:17.146 --> 00:19:21.926 Because you've been working with quantum computers in the past? Yes, of course. 00:19:22.466 --> 00:19:27.546 This question is still open how quantum computing would light would work that 00:19:27.546 --> 00:19:29.346 we need to answer to our audience. 00:19:29.346 --> 00:19:36.106 But first, can you take us through the steps that made you with the co-founders 00:19:36.106 --> 00:19:38.766 start QC Design? Absolutely. 00:19:39.706 --> 00:19:45.726 The founding story of QC Design is that we really wanted to push the envelope 00:19:45.726 --> 00:19:49.066 with what one can do in terms of useful quantum computing. 00:19:49.066 --> 00:19:52.206 Computing and this is something which 00:19:52.206 --> 00:19:55.006 we are very proud of right 00:19:55.006 --> 00:19:58.746 now that our software helps every quantum 00:19:58.746 --> 00:20:02.826 hardware manufacturer on the planet potentially to come 00:20:02.826 --> 00:20:06.246 up with the best designs and basically be 00:20:06.246 --> 00:20:10.086 able to build useful quantum computers two three 00:20:10.086 --> 00:20:14.406 four five years faster than without our software so i think that's the big vision 00:20:14.406 --> 00:20:20.706 and i'm really very happy with the team that we've been fortunate to be able 00:20:20.706 --> 00:20:26.526 to put together for the founding and as well as for the core team here in and around Ulm. 00:20:28.026 --> 00:20:35.546 So the type of software that we build, this computer-aided design for quantum computers, 00:20:35.726 --> 00:20:43.626 this type of software needs three kinds of expertise and one is this expertise in error correction. 00:20:43.866 --> 00:20:48.986 So how do you make a perfect near-perfect logical qubit from many physical qubits, 00:20:49.046 --> 00:20:51.246 it needs expertise in hardware. 00:20:51.546 --> 00:20:58.126 So what are all the nasty things happening to these tiny particles of light or tiny atoms? 00:20:58.846 --> 00:21:01.006 And you need basically. 00:21:02.715 --> 00:21:08.015 Quantum software. How can one harness the properties of this hardware using software? 00:21:08.535 --> 00:21:14.515 So in the team we've been able to from the very early days have this combination of expertise, 00:21:15.995 --> 00:21:20.155 and we started to build this software and realized that it's something which 00:21:20.155 --> 00:21:26.655 many many hardware manufacturers care very deeply about and so we really went 00:21:26.655 --> 00:21:31.415 all in on providing this software to so many of these hardware manufacturers. 00:21:32.335 --> 00:21:38.535 From my understanding, it's actually essential for the survival of those hardware 00:21:38.535 --> 00:21:42.695 manufacturers to be able to really, 00:21:43.255 --> 00:21:51.795 do that, to be an early adopter when quantum computing shows up because otherwise 00:21:51.795 --> 00:21:53.735 they'll they'll be left in the dust, right? 00:21:54.135 --> 00:22:00.435 100%. I think quantum computing is a part of the future of computing, 00:22:00.615 --> 00:22:03.195 a very big part, if not the future of computing. 00:22:04.775 --> 00:22:10.155 And we really need to focus on this task of putting together noisy physical 00:22:10.155 --> 00:22:14.635 qubits into these near-perfect logical qubits, which makes the whole device useful. 00:22:15.115 --> 00:22:22.495 I see. And you said you can be helpful to all hardware manufacturers out there. 00:22:22.795 --> 00:22:25.995 In terms of markets, how many companies were you talking about there? 00:22:26.135 --> 00:22:32.075 In terms of quantum computing manufacturers, there are about 60 or so quantum 00:22:32.075 --> 00:22:35.455 hardware teams right now that are working on building full quantum computers. 00:22:36.095 --> 00:22:41.815 But this is the early days. This is basically the transistor era of quantum computing right now. 00:22:41.815 --> 00:22:49.055 The moment we switch to the pocket calculator era, basically have useful quantum 00:22:49.055 --> 00:22:54.335 computers that solve one commercially valuable problem, then there will be many 00:22:54.335 --> 00:22:55.455 more companies coming up. 00:22:55.535 --> 00:23:00.395 I expect that in the next five to 10 years, this number will go from 60 to way 00:23:00.395 --> 00:23:07.775 more than 100 and including potentially very big companies that are solving 00:23:07.775 --> 00:23:09.875 very important problems for massive industries. 00:23:11.815 --> 00:23:15.135 What comes to mind are the usual 00:23:15.135 --> 00:23:19.995 hardware producers of the pieces that I have here in my new computer. 00:23:20.635 --> 00:23:24.775 How many clients are already using your software? 00:23:25.515 --> 00:23:31.695 So we know that our software is being used in a handful of hardware companies right now. 00:23:32.615 --> 00:23:39.755 Let's say half a dozen or so companies are using it. and this number is something 00:23:39.755 --> 00:23:43.135 that we're working hard on improving in the coming days so we're still a very 00:23:43.135 --> 00:23:47.415 young startup started about two years or so ago and. 00:23:48.978 --> 00:23:52.598 Converted our first customers recently and really now working hard on finding. 00:23:54.018 --> 00:23:58.038 More and more utility for the software more and more people using the software 00:23:58.038 --> 00:24:08.118 do i understand it right that right now talking about old computers here that right now teams who are, 00:24:08.818 --> 00:24:15.718 trying to figure out how a quantum computing cpu could work are now using this 00:24:15.718 --> 00:24:23.138 and over time could Could your software also be used for the companies producing the hard drives, 00:24:23.278 --> 00:24:25.558 the graphics cards, and so on and so forth? 00:24:25.678 --> 00:24:30.318 Whatever is needed in those future configurations of quantum computers? 00:24:31.078 --> 00:24:36.678 100%. That's really the vision. So we want to be able to help our customers, 00:24:36.758 --> 00:24:39.078 these hardware manufacturers, go from end to end. 00:24:39.138 --> 00:24:44.098 Basically, before having to step foot inside the lab, can you design all parts 00:24:44.098 --> 00:24:45.698 of the quantum computer? 00:24:46.298 --> 00:24:49.698 Right now, it is unclear what these different parts are. 00:24:49.878 --> 00:24:55.498 It's still an open research problem. So we have mostly the processor part of 00:24:55.498 --> 00:24:57.258 it where there's a lot of iteration going on. 00:24:57.698 --> 00:25:01.478 But over time, there will also be the memory part of it or the part of it which 00:25:01.478 --> 00:25:09.778 would connect with other quantum computers and do all of this in an error-corrected manner. 00:25:09.998 --> 00:25:14.638 And we hope that one day our customers can use our software to design all of this. 00:25:15.838 --> 00:25:19.798 I see. And one question about. 00:25:20.998 --> 00:25:31.198 As you said, it's not yet clear, but one question about interoperability of those systems. 00:25:31.198 --> 00:25:39.518 Do you believe if there is at one point an economically viable CPU, 00:25:40.038 --> 00:25:45.838 a commercially viable CPU built by a quantum computer company, 00:25:46.358 --> 00:25:52.998 do you believe it would be simply able to be plugged in existing hardware, 00:25:53.298 --> 00:25:58.998 abstracting from the fact that you would need to completely rewrite all the software on it? 00:25:58.998 --> 00:26:03.438 In order for quantum computing to succeed, this is the only way. 00:26:03.678 --> 00:26:10.958 We want to have a world where billions of people can directly program a quantum 00:26:10.958 --> 00:26:12.738 chip like they can a classical chip. 00:26:14.018 --> 00:26:17.358 And I don't know, maybe five to 10 years down the line, do this programming 00:26:17.358 --> 00:26:22.558 without having to learn a programming language, but directly through voice and via AI tools. 00:26:23.078 --> 00:26:25.838 This is the world that should be, 00:26:26.724 --> 00:26:29.904 I also think that there are many smart people working very hard on this. 00:26:29.964 --> 00:26:37.384 There are several top companies, many startups, many bigger companies all working on making this happen. 00:26:37.584 --> 00:26:43.664 How do we make sure that a smart programmer who can code for a CPU today, 00:26:43.744 --> 00:26:45.904 for a classical semiconductor chip today, 00:26:45.984 --> 00:26:51.124 can also learn very quickly how to code for a quantum computer and basically 00:26:51.124 --> 00:26:56.464 make this power of quantum computing available to everyone. 00:26:56.744 --> 00:27:02.324 We've been talking now a lot about your company and quantum computing, 00:27:02.524 --> 00:27:09.224 which actually is, I would say, a deep, if not deep, deep tech startup here. 00:27:09.584 --> 00:27:14.204 Do you think it's possible to build something like this really lean? 00:27:14.844 --> 00:27:17.404 100%. This is something we care a lot about. out. 00:27:18.424 --> 00:27:22.884 Traditionally, people think of lean startups as being kind of opposite to deep 00:27:22.884 --> 00:27:25.024 tech startups. So what's a lean startup? 00:27:25.344 --> 00:27:29.104 This is the standard playbook for, for example, software startups to succeed. 00:27:29.264 --> 00:27:32.384 You build a product, put it in the hands of the customers early on, 00:27:32.444 --> 00:27:35.804 even if it embarrasses you, you learn from this and then build again. 00:27:35.944 --> 00:27:40.824 So the build, measure, iteration loop is at the core of a lean startup. 00:27:41.244 --> 00:27:46.644 And deep tech startups, to put quite simply, they are taking ideas that existed 00:27:46.644 --> 00:27:51.864 so far in a research group in a lab and putting it in the hands of the customers. 00:27:52.104 --> 00:28:01.444 And I think that these two ideas are actually much more compatible than we've thought so far. Why? 00:28:02.204 --> 00:28:05.284 Firstly, because most of the people working in deep tech startups, 00:28:05.464 --> 00:28:10.504 especially the core team, the first 10, 20 people, they often come from science 00:28:10.504 --> 00:28:15.264 and engineering and are so used to the scientific method where you make a hypothesis, 00:28:15.264 --> 00:28:17.344 You go to the lab, test the hypothesis, 00:28:17.564 --> 00:28:20.924 come back, change the hypothesis if the experiment was not successful. 00:28:22.926 --> 00:28:27.186 I think a lean startup is nothing but applying the scientific methods to building a great company. 00:28:27.626 --> 00:28:33.586 So once this kind of mindset is established in a deep tech startup, 00:28:33.726 --> 00:28:38.326 then I think it's very easy to have everyone on the same page about building something lean. 00:28:39.346 --> 00:28:44.106 I can say that for us, this is an experiment that we've been working towards. 00:28:44.106 --> 00:28:47.226 Words. We don't know whether it's successful yet, only time will tell. 00:28:47.326 --> 00:28:54.646 But we focused from the very early days in building things in a matter of days, 00:28:54.866 --> 00:28:58.466 even if it was slow, even if it was, 00:28:59.806 --> 00:29:05.246 90% of the way there, not 100% perfect as a product, and showing it to customers. 00:29:05.506 --> 00:29:10.406 And our customers, being architects and researchers in quantum hardware manufacturers, 00:29:10.406 --> 00:29:17.326 We were so lucky that everyone was extremely open about sharing what are their 00:29:17.326 --> 00:29:21.646 pain points and what is it that is really exciting about such software. 00:29:22.386 --> 00:29:27.066 So for example, a visualization tool that we hacked together over one weekend 00:29:27.066 --> 00:29:33.826 to scratch our own itch ended up being one of the more interesting things for customers. 00:29:37.206 --> 00:29:42.446 And this way of working has solved one very important problem for us. 00:29:42.546 --> 00:29:47.446 So as a company, as a deep tech startup, we are taking ideas at the frontier 00:29:47.446 --> 00:29:50.446 of human knowledge in error correction, in quantum computing. 00:29:50.586 --> 00:29:54.866 We are taking that, whether it's research that we do or research that other 00:29:54.866 --> 00:29:58.286 people do, and taking that and putting it into production level software. 00:29:58.586 --> 00:30:03.646 Now, as you can imagine, the frontier of human knowledge is vast and which of 00:30:03.646 --> 00:30:09.546 the ideas, which of the many cool and fascinating ideas are also useful to our customers. 00:30:09.866 --> 00:30:15.466 This only a lean way of working can help. So everyone in the team is really 00:30:15.466 --> 00:30:20.326 excited that now we know for sure that what we are building is actually being 00:30:20.326 --> 00:30:25.166 used by somebody to build a useful quantum computer, which will push humanity forward. 00:30:25.326 --> 00:30:30.586 So I think that the lean way of working is not only, it's definitely not. 00:30:30.586 --> 00:30:36.306 Not antithetical to building a deep tech startup, but can actually help move 00:30:36.306 --> 00:30:37.666 a deep tech startup forward. 00:30:40.466 --> 00:30:48.646 You've been talking about building startups on the edge of human knowledge and technology. 00:30:49.606 --> 00:30:50.766 You've been saying, 00:30:52.345 --> 00:30:59.905 in India, in Canada, and here in Germany in academics, in different levels, 00:30:59.965 --> 00:31:01.925 even as a postdoc here in Germany. 00:31:02.225 --> 00:31:12.585 I was wondering, do you believe that Germany could do more to enable those spinoffs 00:31:12.585 --> 00:31:16.245 out of universities, research institutions, and so on and so forth? 00:31:16.565 --> 00:31:20.865 This is something I care very deeply about. I think that deep tech startups 00:31:20.865 --> 00:31:24.105 truly have the potential to push humanity forward. 00:31:24.925 --> 00:31:29.225 So whether it was the transistor or the microchip or the internet, 00:31:29.425 --> 00:31:33.945 every time somebody took something from a lab and put it into the hands of customers, 00:31:34.125 --> 00:31:37.545 a lot of things changed. Humanity really advanced. 00:31:39.105 --> 00:31:45.885 Deep tech startups do precisely that. And so far, many of the deep tech startups 00:31:45.885 --> 00:31:49.665 that have come up have been very concentrated, traded, for example, 00:31:49.665 --> 00:31:52.585 in the Silicon Valley or a few other hubs across the world. 00:31:53.965 --> 00:32:00.065 And I think this is really a pity because talent is everywhere. 00:32:01.525 --> 00:32:06.825 And to some extent, money is liquid. It can be everywhere. 00:32:07.005 --> 00:32:10.245 So it's not the lack of talent or money that is holding back Germany, 00:32:10.525 --> 00:32:15.605 Europe, or many other parts of the world from basically having more and more deep tech startups. 00:32:16.465 --> 00:32:19.585 It's something else i feel like there is a 00:32:19.585 --> 00:32:22.565 lot of friction in the process of 00:32:22.565 --> 00:32:25.785 taking something from the lab and actually 00:32:25.785 --> 00:32:28.965 putting it into a company environment and trying 00:32:28.965 --> 00:32:36.465 to take it to customers and i feel like friction if removed can actually 10x 00:32:36.465 --> 00:32:41.725 the amount of deep tech startups coming from coming from research and academic 00:32:41.725 --> 00:32:46.585 hubs in germany and in the rest of the world So friction can be in the form of, 00:32:46.605 --> 00:32:52.045 you know, it's really tough for a PhD student to spend six to 12 months just 00:32:52.045 --> 00:32:55.845 working on some crazy idea and a program, 00:32:56.025 --> 00:33:00.845 for example, where there is a little bit of money and a lot of freedom that's 00:33:00.845 --> 00:33:05.085 given to young researchers to just explore a startup idea. 00:33:05.485 --> 00:33:07.565 Something like this could really reduce friction. 00:33:08.085 --> 00:33:13.265 Friction also comes in the form of kind of tech transfer agreements between 00:33:13.265 --> 00:33:18.225 universities and startups, and I have to say that the U.S. does this much better 00:33:18.225 --> 00:33:19.425 than the rest of the world. 00:33:20.645 --> 00:33:23.605 So, a simple agreement where. 00:33:25.429 --> 00:33:29.169 The university takes a part of the company so that they are also aligned with 00:33:29.169 --> 00:33:32.969 the success of the company and in return basically offers their full support 00:33:32.969 --> 00:33:34.249 in terms of intellectual property. 00:33:35.089 --> 00:33:38.749 I think a much simpler agreement like this, rather than, for example, 00:33:38.749 --> 00:33:41.989 three different ways in which royalties are charged or things like that, 00:33:42.089 --> 00:33:46.149 this would also reduce the friction a lot. 00:33:46.229 --> 00:33:49.989 And many more people don't have to wait months, but instead in a matter of days 00:33:49.989 --> 00:33:53.169 or weeks can basically spin out of a university. 00:33:53.169 --> 00:33:55.989 I have to say we are very fortunate to be working with the University of Ulm, 00:33:56.029 --> 00:34:01.849 but I think this is not the usual experience that people have. 00:34:02.029 --> 00:34:06.329 And it's really, it can take months for IP agreements to be set up with universities. 00:34:07.169 --> 00:34:12.229 And I think the final part of this friction is coming from the lack of role models. 00:34:12.569 --> 00:34:18.009 So for me, this was Martin Plinyu, seeing this world-class researcher who is 00:34:18.009 --> 00:34:22.629 one of the top cited researchers in quantum information, who has mentored hundreds 00:34:22.629 --> 00:34:26.309 of PhD students, many of whom are professors or leaders in the industry now. 00:34:26.809 --> 00:34:32.269 But also Martin wanted to do more than just academia and Martin set up Envision 00:34:32.269 --> 00:34:38.189 Imaging, which is now a quantum imaging startup that's eight years and the product 00:34:38.189 --> 00:34:40.709 is on the market. So this was really inspiring. 00:34:40.969 --> 00:34:46.869 So the fact that it's possible to do it, that also cuts friction. 00:34:47.029 --> 00:34:51.809 So I feel like we should talk more about people trying to build deep tech startups 00:34:51.809 --> 00:34:54.309 or especially deep tech startups that are succeeding. 00:34:54.549 --> 00:34:59.969 So such small things reduce the friction, make it possible for deep tech startups 00:34:59.969 --> 00:35:04.429 to basically come out of everywhere, wherever there's talent and unlock this 00:35:04.429 --> 00:35:08.029 latent potential that's hiding in the universities right now. 00:35:09.711 --> 00:35:15.571 Amen. Actually, one of the many things we try to do here is really to show the 00:35:15.571 --> 00:35:21.271 movers and shakers of the future, those people who found path-breaking startups. 00:35:21.691 --> 00:35:29.251 And that's at the core of StartupRate.io. That's why we'd also love to show more spin-offs here. 00:35:29.771 --> 00:35:33.291 Thank you. let's go we've already 00:35:33.291 --> 00:35:37.211 drifted into our outlook but um i 00:35:37.211 --> 00:35:43.271 would be curious about a few questions and then we can close this interview 00:35:43.271 --> 00:35:49.691 um how do you envision quantum computing revolutionizing industries and what 00:35:49.691 --> 00:35:55.711 real world problems do you believe we could solve with quantum computing that we cannot do now, 00:35:55.831 --> 00:35:58.751 you've been talking about drug discovery. 00:35:59.051 --> 00:36:02.751 And I've talked to a lot of people around here. 00:36:02.871 --> 00:36:06.471 And most of them tell me with a classical computer right now, 00:36:06.611 --> 00:36:13.491 NDAI, they laid the groundwork for the very fast development of the COVID vaccines here. 00:36:14.631 --> 00:36:20.751 These are other problems down the road that we could also solve with introducing 00:36:20.751 --> 00:36:25.471 like real world problems with the introduction of the more capable quantum computing? 00:36:25.771 --> 00:36:30.051 Absolutely. So you already mentioned designing drugs, designing new vaccines. 00:36:30.331 --> 00:36:33.571 I think these will be the first problems where actually we'll see a real impact, 00:36:33.731 --> 00:36:37.271 but also starting to design materials. 00:36:38.371 --> 00:36:41.251 One important material which we don't think enough about 00:36:41.251 --> 00:36:44.431 is fertilizers and fertilizer production 00:36:44.431 --> 00:36:47.831 consumes about 3% of all energy worldwide 00:36:47.831 --> 00:36:50.731 but if you 00:36:50.731 --> 00:36:53.651 see how plants do it it takes a fraction of the energy of 00:36:53.651 --> 00:36:56.371 how we do it at an industrial scale so a quantum 00:36:56.371 --> 00:36:59.251 computer could perhaps help us understand more deeply how 00:36:59.251 --> 00:37:04.551 plants do it and come up with new ways of making fertilizers that could cut 00:37:04.551 --> 00:37:09.471 the energy consumption of the fertilizer process by a factor of a 10 or 100 00:37:09.471 --> 00:37:14.891 so this is just one of the materials would be cool if quantum computers could 00:37:14.891 --> 00:37:17.351 help us design materials for carbon capture, 00:37:17.531 --> 00:37:22.651 would be amazing if we can have new batteries that are designed by quantum computers. 00:37:22.831 --> 00:37:26.611 All of these are things which inherently quantum computers are good at because 00:37:26.611 --> 00:37:29.851 all of these problems involve simulating materials. 00:37:30.351 --> 00:37:34.231 And it's something which classical computers are not very good at, 00:37:34.271 --> 00:37:39.351 but quantum computers being good at being basically quantum can simulate quantum 00:37:39.351 --> 00:37:40.751 materials very, very well. 00:37:40.891 --> 00:37:48.691 So that's my hope for what quantum computers help us with. A lot of talk is right now about AI. 00:37:51.193 --> 00:37:57.533 There's we've been also talking a lot about the increasing capabilities in quantum computing um, 00:37:58.473 --> 00:38:02.313 do you have like any and please don't 00:38:02.313 --> 00:38:05.633 get me wrong i know forecasts are always difficult especially concerning the 00:38:05.633 --> 00:38:13.053 future um but can you do you have a few guesses a few ideas is how the development 00:38:13.053 --> 00:38:21.833 and capabilities of AI could improve with being on quantum computers and not 00:38:21.833 --> 00:38:25.313 the usual silicon-based computers? 00:38:26.753 --> 00:38:32.293 I think right now, AI is growing very rapidly. 00:38:32.433 --> 00:38:35.733 And this rapid part is where things get interesting. 00:38:35.733 --> 00:38:43.653 Existing so the amount of money time energy spent training a new model has gone 00:38:43.653 --> 00:38:49.833 from tens of millions to hundreds of millions to billions now in a matter of only two to four years. 00:38:51.473 --> 00:38:55.933 There is no reason to believe that this acceleration 00:38:55.933 --> 00:39:01.653 will not continue so we do expect tens of billions or so spent training the 00:39:01.653 --> 00:39:06.893 next models and this will continue we can imagine Imagine companies like Microsoft 00:39:06.893 --> 00:39:13.173 and Google and perhaps other entities spending this kind of money training their new models. 00:39:13.873 --> 00:39:16.393 But what happens when we go to hundreds of billion? 00:39:17.653 --> 00:39:23.053 This is really now pushing it. A trillion dollar model? This is not feasible anymore. 00:39:24.153 --> 00:39:29.253 And there is definitely an upper barrier coming very, very fast. 00:39:29.253 --> 00:39:36.993 So the question is, how do we move away from how things are done right now, 00:39:37.093 --> 00:39:42.833 classical computing, to actually come up with new, 00:39:43.013 --> 00:39:45.053 more intelligent models? 00:39:47.422 --> 00:39:51.842 We need an alternative. So quantum computing is definitely an alternative which could help. 00:39:51.982 --> 00:39:55.362 I think this is a very active area of research right now. Many smart people 00:39:55.362 --> 00:40:00.022 are devoting their time and lives towards understanding how quantum computers 00:40:00.022 --> 00:40:02.982 could actually help AI or actually vice versa, 00:40:03.582 --> 00:40:07.522 and have AI help us build better quantum computers. 00:40:08.222 --> 00:40:11.842 I understand. So they can help each other. 00:40:11.942 --> 00:40:15.862 I see. So as soon as the first quantum computer starts working, 00:40:16.162 --> 00:40:19.762 you can build an AI on it that helps you design better quantum computers. 00:40:20.422 --> 00:40:23.562 That'll be interesting. Yeah, that's really interesting. 00:40:24.042 --> 00:40:29.342 But before we get into that, let us get a tiny bit back to QC design. 00:40:30.242 --> 00:40:37.042 The usual closing questions. Are you looking for capable employees? 00:40:37.762 --> 00:40:41.222 And if yes, do they need a PhD in quantum theory? 00:40:42.662 --> 00:40:46.702 Building software like we do needs all the 00:40:46.702 --> 00:40:49.662 smart and kind people that we can have join 00:40:49.662 --> 00:40:52.302 us on this mission so we are 00:40:52.302 --> 00:40:56.022 definitely looking for people we're looking for people who can write pleasing 00:40:56.022 --> 00:41:03.002 fast code we're looking for people who understand very deeply how qubits work 00:41:03.002 --> 00:41:06.862 and how they talk to each other so I guess those are much more research kind 00:41:06.862 --> 00:41:11.582 of positions I guess even Even a smart and ambitious master's student can definitely 00:41:11.582 --> 00:41:15.462 get to this frontier in a matter of months. I don't think a PhD is needed. 00:41:15.882 --> 00:41:21.822 And I also personally feel like for quantum computing to go from a niche industry 00:41:21.822 --> 00:41:25.642 where there's a bunch of startups and two big companies building quantum computers 00:41:25.642 --> 00:41:29.342 to basically it being a worldwide endeavor. 00:41:29.502 --> 00:41:33.462 We need to be able to make sure that many, many smart people can come together 00:41:33.462 --> 00:41:38.922 and build useful quantum computers. So don't let the lack of a PhD stop you 00:41:38.922 --> 00:41:40.502 from applying to a quantum computing role. 00:41:40.622 --> 00:41:44.922 And if you're interested in building software that helps humanity get to useful 00:41:44.922 --> 00:41:50.282 quantum computing, do reach out to us at jobs at QC.design is the email ID. 00:41:52.267 --> 00:41:56.627 And of course, you need to pay those people. Are you open to talk to new investors? 00:41:57.347 --> 00:42:00.947 Yes, we are right now in the middle of a funding round. 00:42:02.607 --> 00:42:08.747 This is to take us from first customer attraction to a true product market fit in a lean startup style. 00:42:09.307 --> 00:42:13.807 We still have a very comfortable runway, so we're not really desperate for money. 00:42:13.807 --> 00:42:17.887 But we think that there's a massive opportunity and really an undervalued opportunity 00:42:17.887 --> 00:42:24.147 in making this EDA or electronic design automation like software for quantum computing. 00:42:24.427 --> 00:42:30.087 So if this is something that is interesting to potential deep tech investors 00:42:30.087 --> 00:42:36.567 or investors in basically B2B software companies, then we'd love to talk. So happy to talk. 00:42:37.007 --> 00:42:40.187 Ish at QC.design is where you can reach out to me. 00:42:40.827 --> 00:42:45.347 And of course, as always, we link down everything down here in the show notes. 00:42:45.447 --> 00:42:48.307 So there will be your LinkedIn profile. People can reach out on LinkedIn. 00:42:49.067 --> 00:42:55.467 We'll add your email in some crypto kind of way. Then you don't get a lot of spam, hopefully. 00:42:55.707 --> 00:43:01.387 Or maybe we just stick to the LinkedIn profile. 00:43:01.827 --> 00:43:05.627 It was a pleasure talking to you. Thank you very much. Congratulations again 00:43:05.627 --> 00:43:10.327 to your German Startup Award Newcomer of the Year 2024. and best of luck in 00:43:10.327 --> 00:43:14.487 building QCD Design further. Thanks so much. It's been a pleasure. 00:43:15.287 --> 00:43:18.067 The pleasure was all mine. Have a good day. Bye-bye. Bye. 00:43:23.627 --> 00:43:32.667 That's all, folks. Find more news, streams, events, and interviews at www.startuprad.io. 00:43:33.147 --> 00:43:35.207 Remember, sharing is caring. 00:43:36.080 --> 00:43:48.906 Music.