Scaling the Quantum Ecosystem in the Netherlands
Overview
In this episode of The Quantum Spin by HKA, host Veronica Combs discusses the barriers to scaling quantum hardware and how QuantWare is removing them with the company’s co-founder and CEO Matthijs Rijlaarsdam. QuantWare’s technology platform helps customers build exponentially more powerful quantum computers. The company also runs a fab that produces quantum chips. Matt credits the Dutch government’s early and sustained financial support for the country’s strong quantum ecosystem. Tune in to understand how QuantWare fits into the full stack ecosystem in the Netherlands.
Matthijs Riljarrsdam has a background in Quantum Information and did his thesis on the simulation of quantum computers in the Elkouss group of QuTech. He also worked at a non-profit strategy consulting firm. When he is not working on QuantWare’s five-year plan, he likes to go hiking in mountains.
00:00 Introduction to Quantum Spin Podcast
00:35 Meet Matthijs Rijlaarsdam: CEO of QuantWare
00:52 Scaling Quantum Processors with VIO Technology
01:51 Challenges in Quantum Computing
02:06 QuantWare’s Proprietary Chip Architecture
04:42 The Importance of Interdisciplinary Collaboration
07:53 QuantWare’s Quantum Error Correction
12:03 The Dutch Quantum Ecosystem
14:11 Future of Quantum Computing
22:39 Conclusion and Farewell
Transcript
[00:00:00] Veronica: Hello, and welcome to the quantum spin by HKA. I’m Veronica Combs. I’m a writer and an editor here at the agency. I get to talk every day with really smart people working on really fascinating subjects, everything in the Quantum industry, from hardware to software. On our podcast, we focus in on quantum communication, and by that I don’t mean making networks safe from hacking or entangling photons over long distances, but talking about the technology.
[00:00:26] How do you explain these complicated concepts to people who don’t have a background in science and engineering but want to understand all the same? Today I am talking with Matthijs Rijlaarsdam, who is the co-founder and CEO of QuantWare. So QuantWare is a very young company, but you’ve done a lot in the last few years, Matt, tell me about it.
[00:00:49] Matthijs: I appreciate that and thanks for having me, Veronica.
[00:00:52] We started our company about three and a half, four years ago now, and we decided to focus on scaling technologies for quantum processors. It’s a technology we call VIO and it allows us to make much larger quantum processors than those you typically see out in the field today.
[00:01:09] And as part of our go-to market strategy, we decided to productize heavily and to then launch early with what at the time was a completely novel business model, a completely novel concept, namely,
[00:01:21] quantum processors that are available commercially. We were the first to do that, and it allowed a lot of our customers that for the first time built quantum computers using components- you know, they buy from vendors. Before that, the only players that could build quantum computers were
[00:01:36] players that made their own quantum chips. And so that’s why, in such a short period of time we were able to build up such a global customer base. We’re now about 20 countries. We were just able to ride that wave and really expand rapidly.
[00:01:50] Veronica: Yes. When folks in the industry started talking about chips I had to think about that for a minute because as you said, I’ve always thought about it like a special purpose-built thing that’s in your office.
[00:02:01] And you can’t just sell one out to someone who comes knocking on your door. But that’s obviously changing. You have a proprietary chip architecture and it’s based on superconducting, is that right?
[00:02:11] Matthijs: Yeah. So what we make is something we call VIO and I, the way you can think about it it’s a physical stack of chips, if you will, and it resides just above the qubit chip.
[00:02:22] And it solves the problems that you run into if you scale superconducting qubits in particular. You can roughly summarize those as threefold. The first one is fan out, so it means that right now superconducting, qubit chips, they route a signal out two dimensionally,
[00:02:39] so in the same plane as the qubits. You can imagine if you have a two-dimensional plane. So the chip, the edges of the chip are only one dimensional. And so if you add more qubits or if you make the qubit chip larger,
[00:02:52] you’re still running out of space at the edges throughout those lines. Thats sounds silly almost, but it’s a pretty fundamental scaling bottleneck. The second is right now to run qubits, you need all kinds of additional components. Things like amplifiers or filters and those are now separate chips in separate little metal boxes.
[00:03:11] And if you want to go to say, a million qubits quantum computer, you will need hundreds of thousands of kilos of equipment in your dilution refrigerator. And obviously that will not scale. And then the third one is yields. And so this means, you know, you can make a qubit, say with 99% probability you can make a qubit.
[00:03:31] This means that if you want to make a hundred qubit chip, the probability of all hundred qubits working on that chip is 99% to the power of hundreds is zero. This scales exponentially bad, and now of course, yield is improving, et cetera, et cetera. If you want to make a million qubit QPU,
[00:03:50] then you need a way to decouple that yield. And so you need to solve all three of those problems. And VIO does this. That’s pretty unique in this space. It’s fully silicon based. We can also produce at that scale. And so in, in one sentence, what this provides in terms of value, what this technology enables you to do is make a large, say, a million qubit quantum computer,
[00:04:12] out of one processor, and then of course you want to network these together, right? So there are all these great companies out there that are making the technology to then link together quantum computers over quantum links.
[00:04:22] And this will be very important to scale these systems, but the larger you make the notes that you’re networking together, the compute power will scale exponentially with that. So not linearly, but exponentially. This is what that provides is, technology platform to build exponentially more powerful quantum computers.
[00:04:42] Veronica: So your background is computer science, that’s what you studied in school. You founded QuantWare with Alessandro Bruno, and he is the physicist in your team. How did you two meet up and start working together?
[00:04:53] Matthijs: By now we, of course, we have a team growing to about a hundred people this year.
[00:04:56] I’m still uh, very underrepresented. As a non-physicist we have PhD professor level physicists mostly of course, as the company grows, smart people from other disciplines as well.
[00:05:11] So my background, like you mentioned, is in computer science. I did a master’s, more in quantum information science. So that’s more, let’s say the theoretical computer science side of quantum computing. I was writing a thesis on the simulation of quantum computers.
[00:05:26] And while I was doing that, through a mutual friend was introduced to Alessandro, to Al, my cofounder, who is arguably the best quantum chipmaker in the world. He was lead of the DICarlo Lab, so he made all kinds of amazing stuff, including some of the world’s first error correction experiments. First quantum computer in the cloud in Europe, quantum chips of Intel.
[00:05:48] So, really amazing stuff. Yeah. And we hit it off and you know had this brilliant invention,
[00:05:53] or brilliant idea on how to scale quantum computers and we were complimentary and we got started basically.
[00:05:58] Veronica: I was at the APS Global Physics Summit in March and I was talking to a few computer science folks and they were saying that, the quantum industry needs more computer science people just to bring everything we’ve learned about building a software stack or a hardware stack or everything we’ve learned over the last, 30 years to the quantum industry to get some of those scaling techniques or just that wisdom into this industry.
[00:06:22] Do you, what do you think of that assessment?
[00:06:24] Matthijs: As a computer scientist, I have to agree. But I would say, I think quantum is interesting exactly for that reason that it’s this intersection of the cutting edge of multiple disciplines as we go to larger and larger systems.
[00:06:39] A couple of years ago, still those worlds are pretty far apart, right? I would be called like, what physicists would call a theorists, right? And those are sometimes on, on almost a different plane. But as we get to larger systems, those worlds get closer and closer together.
[00:06:53] I think that is a very interesting time and place to be.
[00:06:59] What is very interesting is that you have all these people with very deep domain expertise for their particular angle. And so when we started a company, I was allowed to ask a lot of dumb questions on the physics side because, you know, I cannot be expected at the time to, to know some of those things, right?
[00:07:17] Veronica: Yes. For certain
[00:07:17]
[00:07:18] Matthijs: But that allowed me in my role to like to boil things down to the basics, which then in my role as CEO, allows me to explain it to investors, to other stakeholders, so I think that’s actually super super valuable. And vice versa as well, right? Like I could reason about, where might some of the compute advantage be and where might it not be?
[00:07:37] And I think that’s one of the reasons that this is such an interesting field is the cutting edge of all these different disciplines coming together. Yeah, that makes it super exciting to work in.
[00:07:47] Veronica: Yes. Yes. I totally agree. There’s so much knowledge and expertise and just like you said deep, deep experience.
[00:07:53] I know that you also recently announced plug-and-play error correction, which I have to say is a very bold statement, it feels like. So tell me about QuantWare’s work in that space.
[00:08:02] Matthijs: We recently announced our A line QPUs. So the first in that line that we released was Contralto A and the A is standing for algorithm.
[00:08:12] And so Contralto is our medium-sized QPU at the moment. Of course, hopefully soon it’ll be our small-size QPU but it’s a the Contralto-A is a 17-qubit device with tuneable couplers, meaning that between every qubit connection, there is a, yeah, what’s called a tuneable coupler which you could roughly see as another qubit, but it means that you can turn this connection on and off. Very simplified.
[00:08:38] This leads to higher algorithm performance. The way we designed this device is to enable quantum error correction buildup. So you, with Contralto-A, you can build a distance, three surface codes on a computer, you know, if that brings a bell, meaning that it’s large enough to do that, it’s about twice as large as the closest competing solution.
[00:09:03] And it costs about half. It’ll be a pretty attractive product. What this will allow people to do is to because , we will roll out in the near future, we’ll start rolling out larger and larger chips in this algorithm line, and we will scale that quite rapidly.
[00:09:17] Thanks to VIO because again VIO enables us to scale more rapidly than others.
[00:09:24] The customers we are helping with the A line, it’s not just this one single processor, it’s a first step on a roadmap towards state-of-the-art quantum error correction systems of a scale that is beyond what you can currently make.
[00:09:40] , I think the PR you’re referring to is the uh, the joint product we’re doing with QuantrolOX which allows
[00:09:48] scientific customers to quickly tune up their device and automate that as well, which again, lowers the bar significantly to getting to these, what are still state-of-the-art experimental levels. We are really an ecosystem player. We’re creating this ecosystem around this algorithm line that just makes it much, much easier to build state-of-the-art, quantum computers and then scale towards error correction, larger and larger error correction experiments thereafter.
[00:10:15] Veronica: Wow. Wow. It does sound like a good year coming up for QuantWare.
[00:10:20] Matthijs: We certainly think so.
[00:10:23] Veronica: So you mentioned these connectors between qubits.
[00:10:25] Does that provide a form of all-to-all connectivity? Because I know super-connectivity is not necessarily known for that, but I know that is an advantage when you’re trying to, as you say, run deeper algorithms or gates.
[00:10:36] Matthijs: So it doesn’t so superconducting qubits, of course they, the qubits themselves reside on chips, right?
[00:10:41] So those are two dimensional. Now there are techniques and papers that allow you to create long distance connections. That is something that VIO enables. So we will be able to do that. I do think it’s important to point out that all-to-all connectivity is very nice if you have a small system. Okay. As you get to larger and larger systems, again, computer scientist in me talking, as you get to larger and larger systems there’s overhead to mapping all-to-all connectivity to a two-dimensional plane, but it doesn’t scale so badly.
[00:11:11] In fact, as you know, as long the qubit count keeps increasing exponentially,
[00:11:14] you’ll be able to afford not having all-to-all connectivity. Atom based platforms like, I don’t know, neutral atoms or trapped ions, have super high coherence times.
[00:11:22] They have great fidelity, they have all-to-all connectivity. All of that is super important and great for, let’s say these more near-term applications, right? What we are focused on is really very large scale, utility scale, million qubit systems. That is what we are focusing on because that’s where our advantage is with VIO that we can get there quicker. . So what matters is having a lot of qubits and having very fast operations because if you’re doing a large problem, you want to have very deep circuits.
[00:11:50] And if your operations are then too slow, you can in theory calculation, but in practice it will still not be economical.
[00:11:58] Veronica: Okay. So it’s more of a near-term advantage than anything that would scale long term. Yes. All right. Excellent. So you are based in the Netherlands and there is a lot of
[00:12:07] great work and a lot of energy in the Netherlands and in the quantum ecosystem. How does QuantWare fit into that?
[00:12:13] Matthijs: The focus area of the Dutch quantum ecosystem is in Delft. And so we are located in a building where two other quantum companies are.
[00:12:23] There’s one literally a hundred meters to my left. There’s one a hundred meters to my front. We are a spin out of the DICarlo lab, as I mentioned, so is Orange Quantum Systems, so is Qblox. So like a lot of these companies here, the founders know each other
[00:12:38] and most companies started around the same time.
[00:12:41] So we all go way back and then there was this national program called Quantum Delta NL, 6 15 million from the Dutch governments that really supported this ecosystem in its early days. And so I think that combination, the consistent investment into q-tech for the last 10 years,
[00:13:00] then all these founders spinning out and then this support from the government at exactly that moment coming in also, not just for a year, but for longer periods, being very consistent there. I think really that’s what created all of this healthy and diverse ecosystem.
[00:13:13] Veronica: How do you think about these announcements from these bigger companies?
[00:13:17] It still feels like very early days. Are we really ready to just start churning out QPUs at scale? I’m curious about your perspective on that.
[00:13:25] Matthijs: One announcement isn’t the same as the other, right?
[00:13:27] Those announcements differ. Focusing on the Willow announcements. Because I think that’s one that really kicked it off. It’s a fantastic result. It is extremely impressive. Our first angel investor and scientific advisor is Professor Rami Barends who was one of the authors of the Google Supremacy paper.
[00:13:45] And has since left Google, but this is basically continuation of that paper, showing it even larger still. And of course we all know how important that paper was.
[00:13:57] At the same time, it shows also why VIO is so important because, even Google was able to double the number of qubits in five years. If we keep going at that rate it’s gonna take too long to build quantum computers.
[00:14:11] Veronica: Do you think of more building your own quantum machines or do you really see your components being used by other companies to scale up and solve error correction, as you mentioned?
[00:14:20] Matthijs: Our mission statement is to accelerate the quantum computer and the way we do that is by developing VIO and making that the scaling standard. So in one sentence we want to get a monopoly on scaling architectures with VIO and so the way we do that is to get VIO in, in everyone’s system, which means that we will never be a system integrator.
[00:14:40] We are a supplier to all these big companies, but also these smaller companies that are building quantum computers and selling those. The way we bring VIO to market is in these fully packaged products, these chips that you can see on our website, but also foundry services, where third parties that design interesting qubits or that create, novel coupler concepts, that kind of stuff,
[00:15:02] they come to us to fabricate their chips and because we can also offer VIO, we can make much larger chips than they could elsewhere. And the third one, something we’re not doing yet, but we’ll start doing, so somewhere in the next couple of years is packaging services where we package the chip of a third party customer in VIO so they can make those much larger.
[00:15:23] That’s the way we’re positioning the company and you can clearly see this drive of specialization that we assumed was gonna happen and then kicked it off by releasing the first commercially available quantum processor. And now you’re seeing all these pure play system integrators popup.
[00:15:36] So our customers are building quantum computers with our chips and then selling those to third parties. I think that drive will not change and that this drive will in fact accelerate..
[00:15:46] Veronica: So it seems like these decisions and this infrastructure is really customers seeing the need for that and the abilities and resources within the industry seem to be developing pretty quickly then, do you see it that way?
[00:15:58] Matthijs: Yeah. I think that’s also part of, you asked about the Delft quantum ecosystem part of the strength here,
[00:16:04] a lot of these companies spun out of the same group, and so they all picked a different part of the stack, right? Because the control hardware guys spun out Qblox and the chip guys spun out, my co-founders spun out QuantWare et cetera, et cetera. So the Delft ecosystem took a value chain approach.
[00:16:20] For the chip making part, that’s even more extreme. If you ask, the number of qubits increases exponentially the complexity to make them increases. Therefore, that really drives specialization very strongly. So then what is the challenge is to make sure that the rest of this value chain is also there, right?
[00:16:38] Matthijs: That you can build a full stack system using components you can buy. We were kind of lucky because you could buy all of the components except for the quantum chip. So we completed the value chain and therefore this whole revolution kicked off. That being said, as we scale systems, there’ll be other parts that require more specialization, require more developments.
[00:16:56] Creating this ecosystem of products to lower that barrier and go faster and then get to the state of the art and beyond using our products, that is something we are constantly doing.
[00:17:06] Veronica: Right, right. Yes. I think of this in terms of layers of abstraction, and I know that has a very specific meaning in computer science, but just in terms of how much you have to know about how it works at a chip level. It feels like you’re helping move us up that layer of abstraction much faster.
[00:17:24] It’s not like low-code platforms, but I see it as a similar, like a tool to help you start doing interesting things faster as opposed to just putting together the building blocks.
[00:17:34] Matthijs: I think that’s a super good analogy. And I think that’s exactly right.
[00:17:38] That’s also what excites me. If you think about the way classical computers started, at first, everyone was literally programming the bits, right?
[00:17:45] And then, as these systems get exponentially more complex, you need to add more layers of abstraction because, if you have trillions and trillions of bits, you’re not gonna program them bit by bit. And you’re seeing that happening right now, right? So we started with a five-qubit quantum processor, and our customers wanted to make their own software to tune them up, that’s what they wanted to do back then. Right now, our customers want to build systems that run error correction.
[00:18:09] Our products are getting more advanced and if we’re able with a partner to add an abstraction layer in between, you know, that lowers the barrier to entry. More closely for us, we designed our first quantum chips by hand still because they were small enough that you could do that.
[00:18:23] And then Qiskit Metal, for instance, from IBM came around, which automated large parts of it. And then again, as these chips increase in complexity, this software layer to design the chips will become more important. And the same thing happened in semicon, where the first chips were almost literally taped out, these design companies are billion dollar gigantic companies developing the software to design semicon chips. And you can see all these parallels and that’s super exciting.
[00:18:48]
[00:18:48] Veronica: So in addition to designing the chips, you actually produce them as well, is that right?
[00:18:54] Matthijs: We fabricate our own chips.
[00:18:55] Veronica: That seems very impressive.
[00:18:58] Matthijs: Thanks. I think an important nuance to make there is that if you build a semicon fab today, you need $10 billion. If you need a quantum fab today, you need a lot less money because the volumes are much smaller, right?
[00:19:09] We don’t need to make a hundred million quantum chips today. We’ll need to do that someday. Yes. And we think pretty soon. But today that’s not the case, which means that your capital requirements they’re not super low, but we don’t need $10 billion.
[00:19:23] Veronica: Got it.
[00:19:23] Matthijs: To your point, that’s also an area that we have some of the world experts in and it’s super exciting to learn from them. To give an example, we recently had Scott White join as our Chair of the Board. He’s the founding CEO of Pragmatic, left the company as CEO a little while back and then joined us as chair. The search we put out was we want someone that has built a chip making company from scratch. And that last bit is important, because if you have someone that has built a semi-con fab of $10 billion, then he or she’s not gonna understand the scales and the problems and that kind of stuff. Yes. And so our headhunter looks at us, a little bit worried. I think there’s probably one person in the world, maybe two others currently alive that have done that and Scott’s one of them.
[00:20:11] Veronica: Yeah, for sure. I was talking with one of our clients and they were having their chip manufactured and they were saying, this fab only does a few runs a year, and once we get it, we have to test it.
[00:20:22] It could be great, it could be terrible. And I was looking at it from that perspective in terms of the stakes were really high and what if something was wrong and we have to wait another year to manufacture. But thinking of it in terms of the scale that you’re talking about that makes more sense, but still very impressive.
[00:20:35] Very, Very exciting. As I mentioned I spoke to a lot of folks at APS and we talked about different trends in the industry and just things you’re keeping an eye on. Are there any developing projects in the industry or any challenges in the industry that you’re keeping an eye on ?
[00:20:47] Matthijs: I think we’re about to reach a point in the industry where the
[00:20:50] scale and the quality of devices and quantum computers that were, until now basically until this quarter reserved through these really large companies.
[00:21:02] You know that will be available to a much larger group. And I’m curious to see what that will do, in 2026 with the way this field operates.
[00:21:10] And then I think,
[00:21:11] little bit depending on the timing of these announcements, but, towards the end of the year, maybe somewhere in 2026.
[00:21:18] And we’ve seen some of it in this first quarter, more concrete ideas out there on what real utility scale systems will look like, that we can build, fairly soon. And I think that is becoming very exciting. Because these are now becoming almost engineering blueprints.
[00:21:34] They’re not really science projects or, purely visionary renders and papers, they’re almost blueprints. Or in some cases they are blueprints. And I think, we’re about to reach that point. And yeah, and you can already tell that in, in some places that’s really accelerating the field.
[00:21:51] Veronica: Yes, I think everyone has a milestone and this is important and will help us in the future, but it feels like at least at this point, the momentum is real. And it is, it’s consistent and speeding up. And I think the press always looks for the lightning bolt moment or this is what makes everything different from here on.
[00:22:08] But I think it is that accumulation of like you said, expertise and experience and blueprints that really will make things go much faster in 18 months to a year.
[00:22:17] Matthijs: And to be clear , it won’t be here next year. But it will be here much sooner than people think.
[00:22:22] So the picture is getting more clear. And I think that’s the right way to think about it still, a long term view. A good analogy is that you often hear in startups that people overestimate what they can do in the short term and they underestimate what they can do in the long term.
[00:22:35] I think that applies to this field very much.
[00:22:38] Veronica: Yes, I think so too. Matt, thank you so much for your time today. I really appreciate you sharing QuantWare’s work with us and best wishes for a great year.
[00:22:45] Matthijs: Likewise. Thanks so much.
Host Veronica Combs is a quantum tech editor, writer and PR professional. She manages public relations for quantum computing and tech clients as an account manager with HKA Marketing Communications, the #1 agency in quantum tech PR. You can find them on X, formerly known as Twitter, @HKA_PR. Veronica joined HKA from TechRepublic, where she was a senior writer. She has covered technology, healthcare and business strategy for more than 10 years. If you’d like to be on the podcast yourself, you can reach her on LinkedIn, Veronica Combs, or you can go to the HKA website and share your suggestion via the Contact Us page.
July 17, 2025
Leave A Comment