Yuval Boger, Chief Commercial Officer at QuEra, is interviewed by guest host Jack Krupanski. Yuval shares his journey into quantum, discusses the recent Google investment, QuEra’s roadmap, the importance of government support, and the potential for a “quantum winter.” Jack and Yuval also discuss the evolution of quantum computing as part of the broader HPC ecosystem, expectations for production-scale deployment, challenges like talent acquisition, and much more.
Transcript
Jack: Hi, this is Jack Krupansky interviewing Yuval Boger. So, Yuval, who are you and what do you do?
Yuval: My name is Yuval Boger. I’m chief commercial officer of QuEra. QuEra makes quantum computers using neutral atom technology. At QuEra, we believe that neutral atoms are the best way to get to where we all want to, large scale fault-tolerant quantum computers. And we know that QuEra is both the scientific and commercial leader in neutral atoms. And my job as chief commercial officer is to help customers make progress in their quantum journeys.
Jack: Okay, great. One question that came up in my head was that, how did you get started in all of this? Not what was the moment you first heard about quantum computing, but when it was your lightbulb moment when you said, aha, this is something I really need to pay attention to.
Yuval: Throughout my career, I was always attracted to leading-edge technologies. For instance, I ran for many years one of the virtual reality pioneers. We made super high-end virtual reality goggles. I then worked at a wireless power company that can charge your phone over the air safely. And so quantum, once I heard about a quantum opportunity—at that time, it was in my previous company, Classiq—it felt like a natural thing. I have a master’s in physics amongst my other sins. So that sort of combined my love of physics and my passion for sales and marketing.
Jack: Okay, great. And at that same moment, or was it a later moment that you decided that not just that it was interesting, but that you needed to actually, it was for you to get involved in, not just to watch, but to be a participant?
Yuval: Yeah, I mean, it’s always these new technologies. One sort of unserious test that I take to figure out if the market is right for me is to look at the Twitter handles. When I was doing VR, I was VRguy. And when I was doing wireless charging, I was ChargeGuy. And then when I realized that QubitGuy was available, that’s where I figured that quantum was sufficiently early stage so that that would be an industry that would be a good fit for me.
Jack: Okay. So just to recap there, it wasn’t so much that quantum was ready, but that it was early. If it was ready, then that would probably be too late for you.
Yuval: Yes, I spent some time running a cybersecurity company, and in cybersecurity, it felt like if you don’t have 20 years of experience, then you’re nothing. And so I never liked that feeling. So I was always attracted to emerging industries. So yes, quantum was early—or is early. And that’s one of the things that I like about it.
Jack: Great. Maybe you could just highlight some of the high points of the quantum parts of your career. Things that were most satisfying and things that maybe were boring, but necessary, versus things that were exciting but didn’t necessarily have immediate financial value yet still made a difference in feeling like you’re getting somewhere in quantum.
Yuval: I think there are many moments of both types. We do a lot of different things every day. Some are more mundane, blocking and tackling, and some are more exciting. I think as a commercial leader, the sales part is always exciting when a customer agrees to sign a deal. When we won the award for the UK national quantum program, when we closed a computer sale to Japan for AIST—that was a fantastic moment. Just a week ago, although I did not lead that effort, when Google decided to invest in QuEra, that was very gratifying. But then many, many other points. One of the things I do enjoy about this market is, as we are doing right now, I run a podcast. I used to be QubitGuy when I was at Classiq, and now Superposition Guy at QuEra. I get to meet all these fascinating people and ask them questions that I would be personally interested to learn the answers to, hoping that’s representative of what my audience would like to hear too. And so it’s never boring in quantum. It’s never boring to try to write a new piece of software or to look at a new computer. We had a video posted on the Nasdaq tower in Times Square when Harvard and QuEra published what I think is a landmark error correction paper—that was a lot of fun. We had a survey a few months ago where we gave out these 2,000-part LEGO models of quantum computers. A lot of work went into designing that, preparing that, and then sending it out. Both the response to the offer and the survey results were truly interesting. So I’m fortunate to have many high moments in my work at QuEra.
Jack: Okay, great. Now you did mention Google, so let me add a question about that. Could you just briefly summarize what the Google deal was about? I mean, I read your press release and I saw a tweet from Google, but I didn’t see a press release from Google. So I’m not sure what Google thinks they’re getting out of it. Basically, the simple question is, what does QuEra think they’re getting out of this other than just the money? Because it’s a strategic investment; that means there should be something more than finance involved.
Yuval: So we’re certainly getting the money, and we’re also doing some collaborative technical work with Google. Even though Google is using superconducting qubits and we use neutral atoms, there are certainly areas of mutual interest, whether in error correction or others. We also value the recognition. I mean, Google could have invested in many other companies, yet they chose neutral atoms and they chose QuEra. I can’t speak to their motivation; I don’t speak on behalf of Google, but we’re very gratified both by the actual investment and what it signals in the market.
Jack: Okay. Do you have any kind of a timeframe for when you might expect to see some results come from the deal?
Yuval: Now that we have the cash, you can start seeing us hiring even more people, buying even more equipment, thinking about expanding our offices so we can build more systems and, of course, house the people that we hire. How soon the technical output will be—that is still to be determined.
Jack: Okay, great, great. Let’s see, switching gears a little bit or going back a little bit, I should say. Did you have any mentors in your career, either before quantum or in quantum, at any stage? Or were you more self-directed and didn’t need a mentor?
Yuval: I was fortunate to have many mentors, and one conversation that truly changed my professional trajectory was when I finished my master’s. So I did a master’s in physics. It was in fiber optic sensor arrays at Tel Aviv University under the supervision of Professor Moshe Tur. I completed the thesis, defended it, and everything was great. Then I spoke with Professor Tur about what’s next, and he said, “Yuval, if you want to come and do a PhD with me, I’d welcome you with open arms. We have all these amazing problems we could work on, but between you and me, you should go do something else. A PhD may not be the right thing for you.” In retrospect, that was one of the defining moments of my professional career—a professor saying, “Yes, come do a PhD with me if you want, but go do something else.”
Jack: Okay. That reminds me of another question. Looking back at your career, are there any moments where if you had done things a little differently, it would have changed everything for you, hopefully for the better? Are there any regrets, like not getting the PhD? If you had gotten it, would it have mattered? Are there other moments like that?
Yuval: I don’t regret not getting the PhD. I know my mother would have been proud if I did, but now my two sons are pursuing PhDs, so maybe we’ll make it up this way. I certainly have regrets on many things I did or didn’t do in my career. They usually deal with people—could I have behaved better? Could I have been more gracious? Could I have counted to 50 before responding? I hope that now in my 50s, I’ve mellowed a bit compared to things I might have done more instinctively in my 30s. I can do better these days.
Jack: Okay, great. I hate to ask this, but it seems like a pro forma: were there any particular low points or extreme challenges in your career that either you had to slowly recover from or that served as learning opportunities, to put it politely?
Yuval: Although I didn’t plan it this way, I was always the startup guy. I ran software companies, worked in virtual reality and cybersecurity, and was CEO of an ag-tech company for a while. In startups, they say either you succeed or you learn a lot. Sometimes we succeeded—one company I was part of went public on NASDAQ, and that was a great moment. Other times, I’ve learned a lot. I think the hardest moments were probably when you have to tell people, “We’re not going to continue; we didn’t make it; we’re going to have to shut down this company. I know you’re all great people and you’ll find jobs quickly, but as CEO, I failed you. I wish I could have done better.” These were always the learning opportunities. For me, it’s always easier to learn from failures than to learn from successes.
Jack: Okay. Maybe that topic actually brings up an item from the news. Do you have any brief thoughts or a reaction to the demise of Zapata?
Yuval: I know that there are a lot of good people who worked at Zapata, and hopefully, some of them will come and work for QuEra. But this is the nature of an industry—some companies succeed, and some don’t succeed as much. It’s just life. I know Zapata made a lot of contributions to the field, and I’m sorry they had to shut down. I hope all their people will recover and find even more exciting jobs.
Jack: And hopefully, it’ll be a learning opportunity rather than an absolute failure.
Jack: Okay. Now, that also brings up the question of, well, is this just a one-time thing with Zapata, or is it a hint of things to come? I’ll ask your general thoughts on the topic we’ve discussed in the past about the possibility of a “quantum winter.” Does Zapata mean there’s a quantum winter coming, or is it, as I said, maybe a one-shot? How do you feel about the likelihood of a quantum winter in the next two or three years? And what technical or business milestones could either prevent or trigger it?
Yuval: I’m personally not concerned about a quantum winter. I see the size of investments being made both by private investors and by governments. We’re seeing real progress in the types of computers, their size, error correction, algorithms, and more. I think most insiders would tell you that the industry is moving faster than they expected rather than slower. There was, at some point, some concern that AI might pull funding away from quantum computing, but that hasn’t proven true in the long run. Some say, “Oh, we do quantum and AI,” but companies are realizing that these are two different horizons. With AI, you might see useful outcomes in six to 12 months, while in quantum, we’re looking at a longer timeframe.
As long as the industry keeps innovating and focusing on delivering value to customers sooner, I think we’re in good shape. I was talking about that paper from Harvard, QuEra, MIT, UMD, and NIST in December ’23. After that publication, customers came to us, to me and said, “We originally thought useful quantum was maybe five years away, but based on this paper, we’ve shrunk our estimate to two or three years.” That led these customers to engage with us in a deeper way, starting to truly prepare for useful quantum computers.
Just to get back to the beginning, no, I’m not concerned at all about a quantum winter at this point.
Jack: Okay. Since we’ve talked about some of these, like you mentioned in the paper, how was it maybe briefly if you could summarize where you are relative to QuEra, to your roadmap that you had published in terms of this year and next year, things on track. Let me just recap what these customers who were thrilled by that paper, what they can look forward to, let’s say, in the next six to nine months.
Yuval: Well, I’m sure we will publish an update to our roadmap in a few months, but unfortunately, I’m not going to do it here and now. Customers are interested in larger computers and certainly are interested in error correction and logical qubits. There are many steps that have to happen for that. With each generation that we promised and with each generation that some of the other companies in the field promised, it gets better and better. One thing I do want to note is that people sometimes come and ask, “Well, the Harvard QuEra paper showcased 48 logical qubits in December 23. Does that mean I can run 48 logical qubits today?” The answer is that there’s a difference between a university experiment and a commercial system. Our existing system, 256-qubit Aquila machine available for two years now on Amazon Braket, is available for public use for over 100 hours a week. That also started several years before as a science experiment that only worked on Tuesdays between 10:07 and 10:09 after three days of calibration. The scientific paper is phenomenal, and now we’re doing a lot of the work to turn this or something like this into a stable, reliable commercial-grade system.
Jack: Okay, let me drill down a little bit. The roadmap that you have published was saying that for 2024, you’re going to have 10 logical qubits. Is that still roughly on track?
Yuval: We are conducting experiments at QuEra with running algorithms on logical qubits. Soon enough, I hope that we’ll be able to share some of these.
Jack: Okay. If I remember right, the roadmap was that for the next year, for 2025, you would hit 100 logical qubits. Was that right?
Yuval: I think the roadmap was actually showing 2026.
Jack: Okay, two years. The question I actually wanted to ask was, but as you know from discussions with me, I’m a kind of a fanatic about 48 qubits. If you have 48 really good qubits with great connectivity, you should be able to start doing some algorithms. The question is, if we go from 10 to 100, but that takes two years, is it possible we could have something like 48 in the middle of there somewhere, or is it just going to be a two-year desert?
Yuval: The roadmap that we published in early ‘24 spoke about 10, I think 30, and then 100. There was actually an interim step between 10 and 100. Sometimes people ask, well, why this number and not that number? Of course, even to your question, you said you want to see 48 really good qubits. Well, what does “really good” mean? The ratio between physical and logical qubits depends on many things, including the encoding and the code distance and how resilient you want that code to be. Even with a system that has, say, 1,000 qubits, you could have different numbers of logical qubits depending on the level of encoding. But yes, I agree with you that somewhere between 30 logical qubits and 100 logical qubits, this becomes really interesting in terms of that you’re passing the simulation threshold of classical computers, and maybe then you could do some truly fascinating things with quantum computers.
Jack: Okay. Let me just jump to some other random questions here. How is the talent pool for quantum these days? Is it getting any easier or harder to find qualified or at least trainable candidates, or is it about the same? Is your roadmap going to be at risk because of talent pool issues, or how confident are you about the talent pool?
Yuval: We see a number of things that are working in our favor. One thing is that many universities have started up quantum-specific programs to train people more about quantum. The other thing is, we’re increasing the visibility of QuEra. The Google investment certainly helps make QuEra attractive in the eyes of candidates. Well, if Google is backing you, they must know something about why they’re doing it, and therefore this is an interesting place. The third thing is that you don’t need just AMO physicists, atomic molecular optical physicists. As we start building these systems, manufacturing more of them, servicing them, writing cloud software or algorithms for them, you need additional disciplines beyond just what you need for the core of the system. The potential talent pool becomes bigger and bigger as the company develops. Not to mention sales and marketing people, social media, product managers, business developers, finance, HR, and so on.
Jack: Okay. Getting back to investment, because you did mention government. My question here is, does government need to take a stronger role with heavy-duty industrial policy and a lot more funding, or have we crossed some threshold where we don’t need government as much anymore and that the private sector can take on more of the burden? Do we have the mix about right? Would you like to see more government involvement, or what?
Yuval: “I’m from the government, and I’m here to help,” right? That’s the scariest sentence. No, but the government is truly helping in multiple ways. By the way, just yesterday there was a ceremony right near QuEra where the state of Massachusetts unveiled an investment in building a quantum center. Sometimes we call it the “quantum center of mass,” jokingly, and it includes funding to build a computer from QuEra that will be housed in that center. So, not just federal governments, but also state governments are making such efforts. I think that from my perspective, I see government as helping in many ways, but with two parts of two extremes of the spectrum. On one hand, financing and making it possible to perform high-risk projects where the commercial viability is not immediately apparent, where it’s not necessarily going to be something truly useful to a business in six or 12 months but serves a longer timeframe.
On the other extreme—and I think this was the case with semiconductors—where the government can potentially guarantee some level of a market for these products. So yes, we want to buy X hours of quantum computing time, and we want this kind of performance, and if you can do it, then great. So we, as the government, are going to guarantee some level of performance. Of course, the government has other roles. The government can encourage workforce development to make it easier to start quantum companies. Then there is the international side. What should the government, what could the government do to make the world a level playing field? So it is easy for US companies to sell overseas, to remove protectionism. We basically believe in fairness, right? So if, say, it’s difficult for US companies to sell into Europe, then it should be difficult for European companies to sell into the US, or even better, let’s make it equally easy for everyone to sell the best type of quantum computers and let the customers choose—not necessarily the government choose—which vendor is best.
Jack: Okay. And you mentioned Europe, and we have this specter of export controls looming and questions of quantum sovereignty and whatever. Just from your personal experience at QuEra—because you worry about Europe too, right? And you have a relationship with Israel and that kind of background—is the international aspect more of a problem or an opportunity in terms of how you plan and think about strategy?
Yuval: It’s certainly more of an opportunity than a problem. And we’re really delighted that we are able to deliver a system to the UK, to Japan, and hopefully to other countries as well. So we’re very, very glad that we’re able to do that. Of course, the picture is not uniform. Some countries are easier to do business with, and some are a little bit more difficult. One of the really nice things we appreciated about the UK National Quantum Program is that we thought they found a good balance between building and protecting local industry while encouraging innovation from the outside. In the project we won, the requirement was that at least 50% of the innovation happened in the UK. But it’s 50, not 100%, so it’s okay to have 49% of the innovation happen in Boston or someplace else, as long as there is a strong UK partner or set of partners that can continue to advance it. I think they found a really nice middle ground, and we would encourage other countries to do the same.
Jack: Okay. Another government-related question: some have suggested the need for an Apollo-style program for quantum computing to kind of get a critical mass, or what I would refer to as a Manhattan Project, like we had for the atomic bomb. Does that make any sense to you? And if so, does the timing matter, or can it happen at any time? Should we do it now, or do you think it’s needed at all?
Yuval: I mean, there are some aspects that would, of course, be very appealing, right? If the government decided to invest more money toward the goal of a quantum computer that does something, that would be fantastic. Whether this program needs to be centralized like the Manhattan Project—well, that depends on who Oppenheimer is in this particular context, right? So yes, I think countries that invest in quantum are doing it for a very good reason, but I don’t know if it should be completely centralized or whether a more distributed approach might be preferable.
Jack: Okay. Well, how about the national labs? They’re already doing a lot of quantum work, and you have NIST as well, which has done a lot of quantum work. Are you working closely with the national labs? Is that maybe good enough so that an Apollo program isn’t needed? If the national labs weren’t performing their current role, then you might want an Apollo program. But since you have the national labs, are they a sufficient kind of incentive to keep the ball rolling in the more esoteric research areas that may not have immediate commercial requirements?
Yuval: We are doing work with national labs, and we’re very happy about it. Just a couple of months ago, NERSC, part of the Lawrence Berkeley lab, announced they had renewed their contract with QuEra, which has yielded several fantastic scientific results. They even opened an award program for researchers to apply to use QuEra machine time for additional results. So, it sounds like NERSC was very pleased with the program, as were we. And we’re also working with additional national labs. So yes, they play an important role in our progress.
Jack: Okay. On a different angle here, do you feel that any of the existing quantum computing startups, particularly venture-funded, are ever going to become the size of, say, a Microsoft or Intel or Google or IBM? Or are they all destined to either die off like Zapata or be acquired by Microsoft, Intel, Google, or IBM?
Yuval: I think there will be some of each. Some companies will fail, some will merge, and some will survive as standalone entities. Part of it depends on the desires of the CEO and management team, and sometimes it’s driven by external considerations—like running out of funding and needing to make a move. Companies are there to serve the shareholders and provide returns, so sometimes there are opportunities that simply can’t be turned down.
Jack: So, in other words, stay tuned. We’ll have to see.
Jack: There was one company that didn’t show up on that list I gave, and I wrote it as a separate question. If you look at AI in particular, NVIDIA is like the big dog. I mean, you may have Microsoft, Intel, Google, or IBM, but NVIDIA is like all of them put together for AI. But NVIDIA has also expressed some interest and has offerings in the quantum area. So, I’m wondering, do you do any work with NVIDIA, and how do you see NVIDIA’s role? Will they be a secondary player, or will they be one of the primary players in quantum?
Yuval: When we look at where customers want to deploy quantum computers, it’s typically in data centers that already have significant NVIDIA infrastructure. A lot of HPC centers view quantum as the next GPU—not replacing the GPU, but as another coprocessor that, like the GPU does things the CPU can’t, the QPU could do things the GPU can’t. NVIDIA is, of course, very prevalent in data centers. So, we work closely with them. For instance, in the deal I mentioned for delivering a computer to the National Institute of Standards in Japan, called AIST, the computer will be installed next to a classical supercomputer focused on quantum. This classical supercomputer will perform both hybrid quantum-classical algorithms alongside our quantum computer and be a powerful machine to simulate quantum circuits before running them on the actual machine. NVIDIA has definitely invested in simulation with CUDA Quantum and other efforts, and I think they play a critically important role. What will they do in the future? You’ll have to ask them.
Jack: Okay. I guess I’m getting down to my last kind of few questions here. I have lots of questions actually, but a few that I definitely want to get in. One is around production-scale deployment of quantum computers or practical, useful quantum computers. Well, first of all, what term do you prefer to use, and how far out do you think that is? Is it the same two years out that we used to say five or seven years ago, or does it feel like it’s closer than two years, or is it still a fair amount more than two years away? And how would you maybe technically define it so it’s not just an abstract hype meme?
Yuval: When I speak with customers, I ask them, “Where are you in your quantum journey? Where are you today? Where would you like to be, say, 18 months from now? And what’s stopping you? Why are you not there already? Is it money, technology, the right partner, or something else?” When people think about buying a quantum computer, the question is, why do they want to do that? We both know that quantum computers today, some would say, are almost useless in the sense that there’s almost nothing you can do on a quantum computer that you couldn’t do on a classical computer. However, governments and institutions understand that quantum utility—sometimes called quantum advantage—is just a few years away, and they want to get prepared with algorithms, integration into HPC centers, workforce development, and so on.
So we’re definitely not at the point of mass-producing quantum computers in their current state, but we do see a demand and we do see that demand increasing year by year. Maybe three years ago, no one—at least no one at QuEra—would say, “Oh, people want to buy a quantum computer.” They would all say, “Oh, they just want to use them on the cloud.” It turns out that they actually do want to buy quantum computers for various reasons. It could be because they want to run jobs that they don’t want to run on the cloud for security reasons, or because they want to use them as a seed for economic or workforce development in-country or in-region, or because they value control—meaning, if they’re starting to have important jobs, they want to prioritize them relative to what others are doing. So, there are all these reasons why people actually are buying quantum computers.
But we’re not yet at the stage where a production line could pop out 100 quantum computers a week. I think we’re still years away from that.
Jack: Okay. And that also raises the question about size in terms of the size of the organization. Now, if you’re going to buy a quantum computer, that’s a big cost. But if you’re just going to use it casually online in the cloud, anybody could do that, right? You don’t even need two cents. But if you wanted to do a production application where it’s running 24/7, what do you see as the total available market? What’s the low end of the market? What’s the smallest organization size that could probably afford to use a quantum computer for a solution at production scale? What kind of IT budget would they need before you would say, “Look guys, you can’t afford this”?
Yuval: You could make the case that if you ask how much it costs, that means it’s not for you right now in quantum, right? But I think more seriously, you should follow the progression of the GPUs and supercomputers based on GPUs—that will give you a good sense of what quantum will be able to do. So, look at organizations that own a supercomputer today. Why are they doing it? How many are there? How much did they cost? By the way, how much power do they consume? Power consumption is, of course, a significant advantage of quantum computers. And then, sort of follow the same path for quantum computers.
Similarly, with the AI explosion, we saw there were suddenly not enough GPUs in the cloud to meet all companies’ needs. There was a shortage of resources, so some organizations might say, “As soon as I think a quantum computer is going to be able to do something useful, I want to make sure I have that capacity in-house and don’t have to rely on standing in line with everyone else.” Quantum computers today take a while to build—whether it’s a year or 18 months or something like that. It’s not next-day shipping for quantum computers anytime soon. So that also plays into the forward planning of organizations that try to extrapolate when quantum computers will be useful, how much quantum computing they need, and when to buy one to make sure it’s ready in time to meet their requirements.
Jack: Okay. I have to ask a question about hype. So in your view currently, is hype a big deal, a big problem, or is hype itself being overhyped? What kinds of hype annoy you the most, or what kind of hype might do the most damage? Or what do you find most laughable about hype?
Yuval: I mean, I’m in the hype business, right? I run marketing for a quantum computer company. But I think things have sort of tapered down. It’s no longer, “Quantum computers will solve every problem tomorrow morning.” People have gotten tired of that, and of course, it didn’t play out that way. Certain hype is necessary—it’s needed to attract investors and to get talented people excited about working at companies. Otherwise, it’s boring, right? It’s very appealing to say, “Come change the world with us,” as opposed to, “We’re just building a better mousetrap.”
Which type of hype do I find most annoying? Probably competitor hype, right? But I think we’re right-sizing the hype these days.
Jack: Okay. Right-sizing the hype. That sounds like a great takeaway.
Jack: You mentioned marketing, and you’re heavily involved in marketing as your main thing. Is marketing for quantum any different in principle than any other tech marketing? Are there special challenges for quantum marketing?
Yuval: I don’t think so. I mean, you’re speaking to a very intelligent audience. You have to understand where they are in their journey and decide which segments you’re targeting. Are you looking at people who need an education on what a quantum computer is? Are you looking at people who run HPC centers and are trying to figure out where quantum fits in? Are you targeting physicists who want to simulate something else? Obviously, quantum marketing has a significant technical component—it’s not exactly like marketing a new beer. But that’s what makes it fun. You have to explain business value, like in any other B2B marketing. You have to relate to and understand where your customers are and where they want to be. You sort of practice empathy: in this case, what does it mean for you, and how does it make you feel? I don’t think it’s any different than marketing virtual reality products way back when I was running a VR company or something else.
Jack: Okay, great. Getting down to my last few questions here, I hope I can sneak a few in. On that marketing angle again, terminology: do you consider quantum computing to be a full industry, a field, a sector, or what? Or should it simply be considered part of HPC, high-performance computing? Is that maybe where it’s going to end up? Or again, is it a whole industry or just a sector?
Yuval: We look at quantum computing as a new type of compute, but it’s not reinventing the computing industry. Just like we went from CPU to GPU, and now you could do other things. The GPU worked in a completely different way—you had to program it differently and rethink how to structure programs to make the GPU effective. There’s probably a bigger jump from GPUs to QPUs, but ultimately, it’s one more way to do computation.
We used to say, you’re never going to run Zoom on a quantum computer. You’re never going to run Microsoft Word on a quantum computer—at least I hope not. So it’s never going to be the end-all, be-all of compute. But used correctly, it can do things you can’t do with other types of processors. So, it becomes part of a larger compute industry and, initially, part of the high-performance computing industry.
Jack: Okay. What’s the biggest customer objection you run into? I mean, is it money, time resources, or that it’s all too new? What’s the biggest challenge for you when you’re trying to have a conversation with a new customer?
Yuval: I think it’s setting expectations correctly. It’s either, “Oh, if quantum is 10 years away, then let’s talk in eight years,” or on the other hand, “Can you solve my huge traveling salesperson problem today?” So there’s always some expectation matching to do. Some people ask, why neutral atoms or why QuEra? And I think we’ve got that answer down pat, right? Very scalable technology, no cryogenic cooling required, more efficient algorithms, natural perfect qubits with no manufacturing defects, and so on. People sometimes ask why QuEra, and we can make a compelling case that we’re both the scientific and commercial leader in neutral atoms. So it’s really about finding the customer at the right time in terms of what they think quantum computers will do for them and whether they’re interested in engaging now or in the next few months. Otherwise, we can stay in touch and talk again in a year.
Jack: Okay. Maybe as my last technical question here, do you have a rehearsed elevator pitch? Someone might ask two questions: first, why do I need to worry about quantum computers at all now? And second, well, what is a quantum computer anyway? Do you have a 30-second, two-minute spiel for those?
Yuval: I guess it changes from time to time, but here’s one: Quantum computers are a new type of computer powered by the principles of quantum physics. Once sufficiently advanced, they’ll be able to solve problems that classical computers can’t tackle in any reasonable amount of time. We see quantum computers as having substantial potential in several areas, including optimization, such as optimizing traffic patterns; simulation, such as discovering new materials for EV batteries, solar cells, or vaccines; and machine learning. Between simulation, optimization, and machine learning, we believe there’s huge potential for companies to gain a very substantial competitive advantage by correctly using quantum computers. We’re here to help in that journey.
Jack: Okay. And now I guess I’m into your standard questions at the end of your podcasts. First, what keeps you up at night? We’ve touched on this a bit, but maybe your final answer on what keeps you up at night—or, how well do you sleep?
Yuval: I sleep well. What keeps me up at night is wanting to get our amazing technology to customers as quickly as possible. It’s not easy. Sometimes we joke that if building quantum computers were easy, we’d have marketing do it, right? But developing, building, ensuring the software is there, figuring out how to deliver and support them to customers wherever they are—it’s a challenging project. Not many companies in the world can do that. QuEra is one of the few that can, but if only there were a way to do it faster, that’s what I keep thinking about.
Jack: Okay. And the next question is, what’s the biggest thing you’ve learned in the past year or so that was most surprising?
Yuval: I think the biggest thing I learned over the past 12 months is the level of interest in the HPC community and from some HPC centers. And, on the other hand, realizing that some still think of quantum as being 10 years away. We’re trying to change that. I remember giving a pitch at an HPC event not long ago where they were talking about classical supercomputers for machine learning, weather forecasting, and material simulation. We put up a slide and said, “Look, we’re doing weather forecasting with Moody’s, demand prediction, materials simulation with NERSC and others, and here’s our paper on machine learning on how we prove that there are data sets where quantum computers can outperform classical ones.” There are some data sets that exist where quantum is better. And we ended by saying, “There’s a small set of problems—right now, really small but growing—where our 256-qubit quantum computer can outperform all your supercomputers combined.”
To me, that gap between what we know quantum can do and where HPC centers currently stand was a big aha moment in recent months.
Jack: And you mentioned some of the AI stuff in there. I was just curious—since you have a lot of physicists at QuEra, right?
Yuval: We do.
Jack: So, how did they react to this Nobel Prize announcement when it was basically all about AI? Did they react negatively at all? Like, “Well, that’s AI, but it’s not physics,” or did they cheer immensely? How did they react?
Yuval: Well, we can’t give a Nobel Prize to quantum computing every year, only some years. So I think reactions were somewhere between amused and curious to learn more. I know you’ve been investigating that, but it’s certainly an event every year when the Nobel Prizes are awarded. There’s a lot of interest in what’s going on and “Oh, do I know this person? Have I met him? Has he been a professor in one of my courses?” and so on.
Jack: Okay. Now, I was going to come to your dinner question, but I’ll add a twist. Let’s start with breakfast. If you could kick off your week on Monday morning with a breakfast meeting with one or more of the quantum and business greats to review your agenda, plan the week, and really dive into quantum, who might you want to have breakfast with as advisors to set the tone for the week?
Yuval: When you asked about hype earlier, I thought you were going into the hypothetical. I’m an amateur violinist, and when I was very young, just starting out in grade school, I thought maybe one day I could be the best violinist among the physicists and the best physicist among the violinists. Then, of course, I learned that Einstein was a very good violin player, so that idea went out the window quickly. So, Einstein would certainly be my first choice for many reasons, including that. And I’ve asked that question by now to maybe 200 guests on my podcast, and I think a hundred of them have said Feynman. So I’m sure they’re onto something—I’d love to meet Feynman as well. But so many others come to mind, too.
Jack: Okay. I meant for the breakfast side, but would you have the same answer for the dinner question, too? Quantum greats, dead or alive, for dinner—is it the same for breakfast or dinner, or does it matter?
Yuval: Yeah, maybe I’d go for a breakfast or dinner buffet so I could meet all of them at once. But yes.
Jack: Okay. Although the way I phrased the question was originally about quantum greats or business greats, is the quantum sector getting to the stage where it needs a fair amount more business acumen to build on the physics, or are we still at the stage where the primary skill needed is physics?
Yuval: Well, I don’t want to talk myself out of a job, right? About two years ago, when I was looking for my next role in quantum, I reached out to QuEra and said, “Look, I look at your website and see the people on the team—you have a lot of very, very smart PhDs, many physicists. Maybe it’s time you have a business person as well.” And that got me in the door, even without a PhD. So yes, I think building systems is great, but selling them, marketing them, and taking care of the customer is also critically important. Without that, you can build the best science in your lab, but if you can’t get it out to the market, then you’ve only achieved half the goal.
Jack: Okay. And I just have to squeeze in one last technical question. I raised this with you at one point—should the central focus be on research physicists now, or do we need more professional electrical and computer engineers who understand computer architecture? So, do we still need more physicists, or is it time to bring in engineers for commercial product development? Where do you feel we are?
Yuval: Well, we do have commercial products, and they’re very good, but I think there are still many physics problems that aren’t fully resolved—and, of course, many engineering challenges that stem from that. The Apollo 11 onboard computer had, I think, 70 kilobytes of program memory, and it got us to the moon and back safely. It wasn’t written in HTML, right? It was highly optimized to make maximum use of limited resources. I think we’re in a similar spot with quantum—delivering true business value requires full-stack optimizations, from application level to hardware level. Oh, you want logical qubits? Which algorithm do we use, how do we encode it, what are the trade-offs, and how do we implement that in the machine? We’re at these early stages, but like Apollo 11, I believe we can get to the quantum moon and back safely.
Jack: Okay. And I probably have two hours more of questions, but I think we’ve gone way over our time. So, I think that’s it for me on questions. Any last words?
Yuval: No, it was a pleasure and interesting to be on the other side of the microphone. We’ll get to the next hour of questions in a few months, maybe.
Jack: Okay, great. That’s good. Thanks for sharing your thoughts with us today, Yuval.
Yuval: Thank you very much.
Jack: Talk to you soon.