Yuval Boger interviews Helmut Katzgraber, common accomplice at 55North and former quantum chief at AWS. They focus on Helmut’s view that probably the most promising near- to mid-term quantum functions are in quantum simulation for supplies, chemical compounds, and pharma, relatively than optimization. In addition they discover European competitiveness, enterprise alternatives throughout the quantum stack, cloud and on-prem deployment fashions, and the workforce and product expertise wanted to show quantum concepts into profitable corporations.
Transcript
Yuval: Good day, Helmut, and thanks for becoming a member of me immediately.
Helmut: Hello, Yuval. Good speaking to you once more.
Yuval: So who’re you and what do you do?
Helmut: My title is Helmut Katzgraber. I’ve had a number of completely different roles up to now, and now I’m common accomplice at 55North, which is a VC fund centered on quantum applied sciences out of Denmark and to the world.
Yuval: And I spotted that I interviewed you two years in the past once you had been at AWS, and I suppose I missed an interview once you had been a professor on the College of Texas, proper?
Helmut: Yeah, Texas A&M, however let’s not speak about that. That’s lengthy up to now. Lengthy up to now.
Yuval: So now that you just’re now not at AWS, what are you able to inform me concerning the quantum world? What have you ever realized about quantum up to now two years whilst you had been making that transition?
Helmut: Surprisingly so much. As you bear in mind, or may bear in mind, I led a crew of scientists that now are being taken care of by Martin Schuetz, who’s a incredible new chief, centered on a) discovering functions of quantum applied sciences. You already know, the place can a quantum machine have influence in the true world, not in toy issues, however be additionally, you understand, sort of increase the bar for quantum by fixing the issues classical in an ideal world, having options working in manufacturing setting of AWS as clients. And never simply through the previous two years, however over the last 5 to seven years, even once I was at Microsoft, I gathered lots of expertise into what this machine can and can’t do. And specifically, which use circumstances are good use circumstances for quantum applied sciences and which of them doubtless are gonna wrestle down the street.
Yuval: And what’s the reply?
Helmut: Nicely, the reply is straightforward. Use quantum the place quantum reigns supreme, you understand, within the quantum world. So from my perspective, if we have a look at close to to medium time period quantum gadgets, doubtless the primary influence goes to be in issues associated to supplies, chemical compounds, pharma, the place you have to do quantum mechanical simulations. There was only a latest paper out I believe this morning or final evening the place a fabric was simulated on a impartial atom machine, and it’s these kinds of workloads which are actually going to be the primary ones that ship one thing that hopefully in some unspecified time in the future classical can’t do. Different use circumstances within the computing space, you understand, machine studying, I might say I don’t wish to hedge my bets but just because, you understand, it’s nonetheless comparatively nascent and there’s nonetheless so much to be realized. And in optimization, I’m comparatively bullish that classical applied sciences for the foreseeable future will outperform any quantum machine.
Yuval: And I do not forget that in optimization you had been a skeptic or some would say a realist, however you’re saying mainly in materials science, that is the place the hope is.
Helmut: Yeah, I might say so. Particularly, you understand, once you’re simulating strongly correlated issues, you understand, like battery supplies, natural LEDs, you understand, sure drugs which have steel natural complexes. That’s the place classical compute struggles and the place quantum applied sciences hopefully will carry worth down the street.
Yuval: Immediately, you’re in enterprise capital and individuals who spend money on your fund need to see returns. Do you assume the returns for enterprise for quantum investments are going to be long run than conventional VCs or are you anticipating the identical timelines?
Helmut: Like my colleague Owen Lozman, the managing accomplice, likes to say, you understand, we’ve unpatient or how does he put it once more? We have now a long-term perspective with unpatient capital. And what he means by that’s that we make investments and there are applied sciences that may carry returns within the typical VC timelines. Consider enabling applied sciences, refrigeration, for instance, one thing that many modalities will want, or software program to compute error correction fashions. Once more, one thing that, particularly now that we transfer from bodily to logical qubits, one thing that everyone’s gonna want. So there are specific areas the place we will see returns in a what I might name significant timeline, you understand, through the lifetime of the fund. And naturally, there’s all the time the one or the opposite unicorn like IQM not too long ago that had us again, which may carry returns sooner and later.
Yuval: You had been born in Austria, proper?
Helmut: No, I used to be born in Peru.
Yuval: In Peru. Oh, and sure, you will have studied in Europe, proper?
Helmut: Yeah, so I used to be born in Peru, lived my first 20, virtually 20 years in Lima, after which I studied in Switzerland, physics.
Yuval: And when you consider, and proper now you’re within the European VC, do you assume Europe can compete with the US and with China on quantum computing? And in that case, why?
Helmut: Sure and no. And I do know this can be a superposition of two solutions, however Europe has incredible expertise. I might say, you understand, the tutorial system is top-notch, among the best, if not the very best on this planet and throughout many international locations within the union. You will have the mandatory ecosystem for quantum applied sciences to thrive in labs, you understand, with beneficiant grants. And the purpose the place it begins faltering, and it’s a fairly leaky pipeline from that standpoint, is relating to taking these concepts from the lab right into a product. And so, you understand, 55North tries to handle this hole. That’s the place it was based with myself, Oren Lozman and Kai Hudek, with Novo Holdings, or sorry, Novo Nordisk Basis and EIFO as LPs, to mainly bridge that hole and take these concepts from the lab out into startups. Now, the problem that we’ve in comparison with the US is that there’s way more capital over there, and particularly development capital. And that’s one thing that the EU is attempting to resolve proper now. There’s this Proposition 28, the place they attempt to harmonize the legal guidelines throughout the union to make it simpler for startups. However it’s nonetheless going to take time earlier than that takes an impact, if in any respect. You by no means know with the union. And so the large problem is that whereas we’ve the good concepts, what we’re missing is the enterprise mindset in Europe. We’re missing the product mindset, and extra importantly, we’re missing the expansion capital that may come down the street. And I believe that, you understand, if we will resolve not less than two out of three, we’ve a very good likelihood of being very profitable in comparison with, specifically, America.
Yuval: Are employment legal guidelines additionally an element? I imply, in some international locations, it’s simpler than others to rent and fireplace. You already know, startups are dynamic, issues change. Do you assume that performs a task?
Helmut: I believe that performs much less of a task. I imply, we work with a bunch of startups in Europe. We’ve invested in a couple of corporations in Europe, like Kiutra or IQM. And it’s much less of a problem, to be trustworthy. I don’t assume that the employment legal guidelines are a hindrance. If something, there’s one thing good. I by no means favored the actual fact within the US that I may get fired any minute for no matter cause, with out a warning. And that nervousness is one thing that comes from the truth that once you take an educational position, you’ve got a sure job safety that’s fairly protected. And so abruptly you’re in a world the place something can occur. And so I believe that for teachers, the European labor legal guidelines are most likely way more friendlier or extra comfy to cope with than within the US.
Yuval: How are you interested by the geopolitics of quantum? You already know, is it, immediately it’s not that simple as an example, for US corporations to promote into the EU. It’s not unattainable, but it surely’s not simple. That scenario may grow to be symmetric someday. How are you interested by that and the way it impacts the fund and the quantum enterprise?
Helmut: So I fully agree with you. There are some challenges. For instance, once we spend money on the US, we’ve to watch out that we’ve only a small fraction of the corporate after the funding, in order that we don’t set off export management legal guidelines. And so there’s a bunch of challenges sort of going each methods, really. My hope is that it’ll get resolved. As I mentioned, the EU is engaged on harmonizing a few of these legal guidelines and making them extra founder-friendly, however you by no means know. I imply, I believe we each know who’s on the helm in america and it’s a finger within the air and see what occurs. You already know, it may go this fashion or that means any minute. However I believe that it’s a threat and it’s a threat that we’re watching rigorously just because, you understand, our mandate is to take a position principally within the Nordics, but in addition throughout Europe and the remainder of the world. And for instance, if a rustic is in an lively battle zone then we’re not allowed to take a position. It would change, however on the identical time, our LPs have given us lots of leeway in order that, you understand, we will make investments worldwide.
Yuval: When you consider the quantum stack from {hardware} and {hardware} that helps {hardware} like dilution fridges that you just talked about, middleware, software program, utility, cloud service, the place’s the worth or the place do you anticipate the worth might be for an investor?
Helmut: That strongly relies upon, initially, on the corporate. We like corporations that assume outdoors the field, that strive issues that no person else has tried, which are going towards the mainstream. And so, you understand, if you consider a startup competing towards one of many huge tech corporations, IBM, Google, Microsoft, Amazon, which have mainly infinite pockets to construct quantum gadgets, you then may as properly say, properly, the startups would simply hand over on the spot. However the actuality is that startups are way more agile. And what this implies is that, you understand, they’ll experiment on some issues that no person else would dare to. And so there’s lots of potential worth in these, I might virtually name them moonshot concepts or one divided by moonshot concepts. Equally, you understand, there’s worth in enabling applied sciences, as I discussed earlier, refrigeration, you understand, software program that a number of corporations may want. I believe it’s just a little early for functions. Possibly in some areas of chemistry, there’s a couple of issues we’ve seen which are fairly fascinating, however I believe that the machines should not there but to ship any sort of industrial influence in an utility scale. They’re nonetheless experiments. And so what we concentrate on is we have a look at the entire stack, prime to backside, backside to prime. We have a look at some full stack corporations, however we additionally have a look at completely different areas of the stack, specifically these that may deal with completely different modalities.
Yuval: Do you assume full stack corporations, placing apart the large corporations, however do you assume full stack corporations is only a short-term resolution as a result of somebody’s obtained to do it? Or do you assume that’s a sustainable enterprise mannequin?
Helmut: That’s a very good query. And actually, I don’t know the reply to it. It’s laborious for me to say if it’s gonna be a sustainable mannequin, you understand. Any non-big tech full stack firm has a comparatively small likelihood of success basically. We’ve had some successes, you understand, there’s IonQ, there’s QuEra, IQM, some corporations which are doing very well and are very properly funded. However you understand, you’ll be able to depend in your each palms the businesses that actually have a very good shot at succeeding versus all of the others which are attempting and finally find yourself pivoting to doing one thing else. I’ve seen corporations that needed to, for instance, construct qubits after which realized that, for instance, their management electronics are much better than what everyone else has, however they’re not good at constructing qubits. So may as properly simply pivot and concentrate on that exact space. And so full stack, we may have these unicorns right here and there, however we’ll see finally some kind of coalescence of the market, the place among the full stack corporations, even the startups, you understand, changing into larger and larger, after which mainly completely different corporations focusing extra on elements that these corporations may want.
Yuval: I wish to speak just a little bit about cloud companies, once more, constructing in your expertise at AWS, and even previous to that at Microsoft. The primary query is how ought to a classical cloud supplier regulate to quantum expertise changing into a part of the computational material?
Helmut: I believe an important factor is to deal with it like, you understand, a brand new compute modality, the identical means that you would be able to entry FPGAs, GPUs on AWS, you’ll be able to even entry Mac minis within the cloud. And on the identical time, the significance although is to actually construct a software program stack in a means that you are able to do hybrid workloads. Fairly often you’ll have the quantum gadgets in a service, you’ve got the classical compute in several companies like on Amazon, you’ve got EC2 versus Braket, you understand, however there needs to be a tighter integration for among the extra necessary workloads. I believe Microsoft did one thing fascinating with Azure Quantum Components the place they mixed AI compute and quantum in an effort to mainly make an end-to-end chemistry pipeline. And I believe there’s lots of worth in one thing like that.
Yuval: And on the sustainability query, and I don’t imply sustainability as in inexperienced vitality, however immediately having a quantum laptop on the cloud may be very enticing as a result of one may experiment and swap rapidly and comparatively inexpensively get into quantum. As quantum computer systems grow to be really helpful, do you see them shifting extra into on-premises? Do you see a 50/50 cut up? How are you interested by that?
Helmut: It’s laborious to inform. I can see many, particularly the HPC facilities, if you consider Juelich, for instance, shopping for these gadgets, making them accessible. In Denmark, which by the way in which, has a incredible quantum technique, the purpose is, for instance, to obtain and construct a machine by QuNorth and make it accessible. So you’ll have some on-prem machines, however I think that lots of them will stay within the cloud. And it has principally to do with the truth that, you understand, it’s not a normal laptop you plug into the wall and that was it, possibly some cooling. It requires much more babysitting for the foreseeable future. So having these machines housed on the producers, for instance, is gonna have some benefits that if one thing goes incorrect, they are often mounted. In different phrases, cloud entry being optimum. However on the identical time, I can think about that sure establishments may have particular causes, be it nationwide intelligence or so, to have these machines on-prem. So I think that in the identical means that HPC clusters exist on-prem and within the cloud, we’re gonna have one thing related. Though you see increasingly more how universities are giving up on their HPC clusters and shifting increasingly more workloads into the cloud.
Yuval: You talked about Microsoft Components, which is a vertical utility versus a horizontal common function laptop. Do you assume that’s the way in which to go? Or what would you suggest to a portfolio firm making quantum computer systems? Ought to they provide it generically or ought to they discover and create a full stack resolution focusing on a selected utility?
Helmut: So it relies upon what the corporate needs to do. If the corporate needs to construct a analysis machine, then by all means, make it generic. Everyone experiment with it. My private take has all the time been, and you understand this, begin with the issue after which construct an answer round the issue. And so, excuse me, should you ask me, I believe corporations ought to say, okay, we wish to goal the simulation of strongly correlated supplies within the space of battery supplies. After which mainly construct the entire product round that. Now you may say, properly, isn’t {that a} very slim use case? The reply isn’t any. There’s just a little worth in it, not solely that, however extra importantly, you understand, by altering a couple of issues, you’ll be able to modify it for different sorts of strongly correlated supplies. And so additionally based mostly on my expertise at AWS, you understand, customized options to troublesome issues will all the time outperform any sort of generic instrument. And so if we wish to see early utility worth, it would make sense to concentrate on use circumstances. On the identical time, provided that the machines are nonetheless of their infancy, I believe it’s too early to try this. That’s why I hold going backwards and forwards. And so as soon as folks experiment, establish potential use circumstances of worth, after which mainly hone the machines in that exact path.
Yuval: You talked about just a little bit concerning the geographical focus of the fund, however simply give me the background. How massive is the fund? What are you investing in? At what stage? Is it simply computing or are you additionally gonna do sensing and communication and what have you ever? So what’s the, you understand, the 2 paragraph abstract of what you’re doing for the fund?
Helmut: So the fund’s referred to as 55North. All of us moved to Denmark to launch the fund. I gave up all the pieces within the US, packed my baggage and I’m now in Denmark. And the purpose, the primary shut was 134 million euros, with the purpose being to concentrate on all kinds of quantum applied sciences and adjoining applied sciences. The fund can develop as much as 300 million euros. We’re in fact in the course of fundraising as you may think. And we’re three companions with complementing experience. Owen Lozman is a chemist by coaching. He labored in chemistry R&D in business his entire life, finally took over Merck Ventures, so comes with the expertise of enterprise capital. Kai Hudek was one of many first workers of IonQ. He has bodily constructed these machines, so he understands {hardware} very deeply. And I carry the speculation, computation, and cloud perspective to the combination. So the cool factor is that, you understand, I completely didn’t anticipate this. Founders actually prefer to work with us as a result of they need board members that perceive what they’re doing. And so that’s, I believe, sort of like the key sauce of the fund. We make investments from pre-seed all the way in which to Collection B, so we all the time hold some cash on the aspect for later rounds. And we additionally focus, you understand, within the Nordics and in Europe, however we’ve a worldwide mandate. And so all the pieces is honest recreation for us.
Yuval: What’s your favourite modality?
Helmut: Good query. Now, earlier than I reply that, I forgot one element. And that’s, sure, we do additionally spend money on sensing, you understand, and networking. So we, as I mentioned, we spend money on something that we really feel may carry worth. Now, what’s my favourite modality? I’m going to placed on the Helmut hat and I’m going to provide you a really non-committal reply. And that’s the one which brings the very best worth for a selected utility. And so, you understand, in order for you quick gadgets, say if you wish to do variational workloads, it’s clear that it’s going to be one thing that’s superconducting, for instance. However they do have their disadvantages. If you wish to have plenty of qubits, impartial atoms are good. And in order for you very prime quality qubits, then ion traps may do your finest. However I believe that no modality has a giant sizable benefit over the others. And so, as I mentioned earlier than, my favourite modality is the one which brings most worth for the issue you’re attempting to have a look at. And so, once more, it is best to use the correct instrument for the correct downside.
Yuval: Workforce improvement has been spoken about typically within the context of quantum and I’m certain as a enterprise fund you’re additionally getting requested about that. What would you suggest a university graduate or a university pupil to review in the event that they wish to grow to be a part of the quantum world?
Helmut: I believe that it relies upon in fact what position. You already know, for sure belongings you want deep physics experience. If you wish to do error correction, then there’s no different means than being a theorist. If you wish to construct gadgets within the lab, then we will bend the principles just a little bit and say, you understand, a mix of utilized engineering or utilized physics may do the trick. I believe these are most likely the clearer roles. What I might like to see extra is a product mindset. And so folks that have product experience, you understand, product managers which are keen to take a second diploma, you understand, like mainly two tracks, one in physics and one in product administration slash enterprise administration, that will be extraordinarily beneficial for many quantum corporations. And that is likely one of the challenges we see most frequently once we speak to startups that, you understand, they’ve nice concepts. And once you say, so are you able to make me a thousand of these? It’s like, oh God, no, I don’t know how. And so, you understand, these hybrid levels, enterprise and quantum, I believe are those that we’d like. To my recollection, and you understand, the one individual I do know that actually purposefully did this and also you might need heard of him is Andre Ochoa, who’s at Strangeworks. He obtained his PhD in physics, then he went straight to enterprise college to get an MBA, in order that he can mainly fill precisely that hole.
Yuval: Hypothetical, and I bear in mind what you answered final time I requested you that. So we’ll see should you bear in mind as properly otherwise you wish to change your reply. If you happen to may have dinner with one of many quantum greats, useless or alive, who would that be?
Helmut: My reply nonetheless stands. It’s going to be Dirac. Not solely do I discover his arithmetic probably the most elegant on the market, however he was a person of few phrases. He’d typically dine alone and I wouldn’t thoughts eating with anyone that doesn’t yap on a regular basis. Copenhagen has lots of nice meals choices, so I wouldn’t thoughts if I may flip again time to satisfy him right here in Copenhagen and simply go to any of the eating places and luxuriate in a meal in peace and, you understand, possibly speak just a little bit concerning the magnificence of arithmetic.
Helmut: In a area that’s altering so rapidly?
Yuval: That was a really constant reply, very very like what you answered.
Helmut: You already know, I’ve his books up right here on the shelf and as I mentioned, it’s probably the most elegant means once you have a look at his math, the way in which he solved issues.
Yuval: Helmut, thanks very a lot for talking with me immediately.
Helmut: Thanks, Yuval, for having me. It was a pleasure.
Yuval Boger is the Chief Business Officer of QuEra Computing.
June 22, 2026
