Daniel Rodan Legrain: Alright, hello everyone, hello Kyle.
Kyle Serniak: Hi, Danny.
Daniel Rodan Legrain: I'm Daniel Rodan Legrain. I'm the Quantum Science Engineering Lead for North America at Qblogs, the sponsor of today's And these five chat series basically brings together series shaping the quantum ecosystems ⁓ all the way across academia, research institutions, and industry. And in previous editions, we've had the pleasure of hosting Will Oliver, Berta Jong, Peter Shore, Steve Girvin, Gretchen Campbell, Chuck Black, David Afshalom, among others across the stack. And today, I'm very happy and excited to welcome Dr. Kyle Cernyak. ⁓ Kyle, you're the assistant group leader in the quantum enabled competition group at the MIT Lincoln Lab, your research affiliate at ⁓ MIT RLE and ECOS, and part of the MIT Center for Quantum Engineering. ⁓ Kyle's work focuses on improving superconducting quantum devices, basically spanning noise, quasi-particles, materials, device design, fabrication, and also on building shared infrastructure for the community, including the Squill Foundry. Kyle, you completed your PhD in Michel de Voré's group at Yale, working on non-equilibrium quasi-particles, and a topic that I would say remains very central to qubit performance today. ⁓ Kyle, again, it's a real pleasure to have you today here. it.
Kyle Serniak: Thanks for the introduction Danny. Yeah, I'm really excited to chat with you today Thank you so much for inviting me to do this and I'm looking forward to sharing Yeah, yeah sharing this discussion with you
Daniel Rodan Legrain: Fantastic. Let's dive in. Before that, for everyone ⁓ watching live, please feel free to post your questions in the chat, and I will bring a few at the end of the discussion. So, Kyle, why don't you start with your own journey? I mean, how you ended up basically working at the intersection of device physics, fabrications, and community infrastructure. So...
Kyle Serniak: Yeah, yeah, happy to. So I started out in this field as an undergraduate. I was completing my undergraduate degree at Florida State University. one of the, you know, a really great resource that exists there is the National Hamadetic Field Laboratory. And I was a physics major and, you wanting to get into research. I had already done some research in the chemistry department, but wanted to make a transition more into physics. With the magnet lab there, I was looking for opportunities sort of therein and got involved with Irinal Kirescu, who was one of the early folks working on control of flux qubits. was a postdoc in Delft doing that. With Hans Moi. And ⁓ yeah, got involved there. And that was kind of where I started getting into microbe engineering and all of the sort of underpinnings of the superconducting qubit field. And was also where I first experienced sort of the big lab kind of atmosphere. ⁓ of the National Ammonautical Field Lab at the, well, it's today, but there in Tallahassee, which kind of pointed me toward my trajectory also in super90 cubits at relatively large research institutions, MIT Lincoln Lab.
Daniel Rodan Legrain: Wonderful. Was there any key moment that made you realize that this is the field you wanted to be focusing on?
Kyle Serniak: I think it was, I think it was just a confluence of things. You know, there was opportunity specifically, you know, to do research related to this field at the Mag Lab and, and, you know, without, without that, I don't know that I would have ended up on this trajectory, to be honest with you. So definitely, you know, a target of opportunity and... And I think I knew that I would stick with it once I got in the lab. I saw all the different aspects that kind of contribute to the field. The fabrication side, the RF design side, the cryogenic side. It felt very dynamic. It felt very exciting. It felt like I would never feel like I'm doing the same thing day after day after day, which is certainly true, I would say, even now. ⁓ I think it was the combination of those things that made it click for me that ⁓ I had already known pretty strongly that I was interested in research at that time, ⁓ but really that this field was the right fit for me was probably around then.
Daniel Rodan Legrain: And here we are now, at the right place at the right time.
Kyle Serniak: You know, ⁓ it's easy to say that everything is merit-based and if you work hard, you get there, but there's definitely an aspect of luck, I would say.
Daniel Rodan Legrain: So talking about that, I mean, you now operate across MIT and Lincoln Lab. How would you actually describe kind of like the value of that model in practice? And what does that environment enable you to do that maybe would be difficult to achieve in just one of the two settings?
Kyle Serniak: Yeah, no, it's a question. maybe for a little bit of context, ⁓ MIT Lincoln Laboratory is a federally funded research and development center that's operated by MIT. So everybody who's a researcher out at Lincoln Lab, we're all MIT employees. And there's been a long history of superconducting qubit work. ⁓ both at MIT starting with Terry Orlando and then at Lincoln Lab with Will Oliver when he was here, along with Mark Gouker, a bunch of other folks that kind of started those efforts 25 years ago at this point, thereabouts. And so there's a lot of history there and there's always been a history of tight coordination and collaboration between the teams out at Lincoln Lab and the teams at MIT campus. So, you know. It's a really great environment to work in. On the Lincoln Lab side, have staff scientists, generally speaking, with a lot of institutional knowledge, people who love what they're doing and have been doing it for many, years. ⁓ And from the campus side, it's an academic group. So you've got students and post-docs with fresh ideas and, you know, just boundless enthusiasm coming through and bringing those folks together, ⁓ you know, it can make for some really exciting research and development. ⁓ So, you know, I love it, to be honest with you. ⁓ It's a great situation to work within. I'm thankful every day that I'm able to contribute to this ⁓ research ecosystem.
Daniel Rodan Legrain: I can say I was fortunate enough to leave it during my own career and I can say it's a really special place and you are definitely one of the key pieces to make it happen and work, Kyle.
Kyle Serniak: Yeah, yeah, Yeah, well, thank you. Thank you for saying that. guess we can break the fourth wall a little bit and say that, that we worked a little bit together when you were a postdoc in the Engineering and Quantum Systems group, Wallerbergs group at MIT, ⁓ working on 2D materials and gate mod qubits and things like that. It was a lot of fun.
Daniel Rodan Legrain: Indeed. ⁓ I wanted to also ask you, Kyle, was there anything, something that maybe surprised you a little bit more when you moved to a, let's say a more engineering driven environment?
Kyle Serniak: being like a laboratory.
Daniel Rodan Legrain: Sorry, Bing.
Kyle Serniak: Like Lincoln Laboratory being the more engineering driven environment. Yeah, there were some adjustments, ⁓ but really, you know, maybe less so from an engineering perspective and more a, ⁓ you know, professional perspective. So Lincoln Lab operates in an interesting space between academia and industry. ⁓
Daniel Rodan Legrain: Indeed.
Kyle Serniak: the way the day-to-day looks is maybe more similar to what you'd expect in industry. ⁓ It's a little bit more structured, little bit more, ⁓ yeah, structured is the right word. ⁓ But we're doing a lot of academic style research. ⁓ so getting used to doing really the same things that I did during my PhD, designing. fabricating, measuring, analyzing data, doing experiments ⁓ in a different setting that was a little bit more methodical is maybe the right way to put it. In a PhD program, you're kind of go, go, go, and you're kind of doing everything yourself. ⁓ In a more professional setting, there's more distribution of responsibilities that makes for really, really excellent work and really methodical work. that's just a bit of a shock, you know, moving from one to the other. And there's pros and cons to both, but I definitely really enjoy the more methodical, the more. which is forward looking, let's say, ⁓ in the way that we structure the work at it.
Daniel Rodan Legrain: Thank you, Kyle. Let's maybe now dive into the physics and maybe zoom into one of the central themes of your work, what actually limits ⁓ qubits performances. So a lot of the field talks about T1, T2, but your work actually goes much deeper than that into the underlying mechanisms, especially non-equilibrium quasi-particles. What makes them such a persistent limitation and why do they still matter so much today?
Kyle Serniak: So ⁓ quasi particles, non-equilibrium quasi particles are interesting because they're just so tightly ingrained with a lot of the interesting and kind of weird parts of superconducting qubits as a hardware modality. ⁓ We often talk about superconducting qubits as artificial atoms, right? There's this picture that they have complex. ⁓ energy level structure and interesting selection rules sometimes between, you know, for transitions between states. Just like, ⁓ you know, a trapped ion or one of these more archetypically quantum kind of systems. ⁓ But the only reason that we can ever really resolve any of that quantum phenomena is if we're in the limit where we can just completely forget about all of the electronic degrees of freedom in these circuits. You know, we write down a superconducting qubit with, you know, inductors and capacitors and okay, there are Josephson junctions, which are a little bit, a little bit non-standard from a circuit perspective. But otherwise we're talking about components that people think about holding in their hand and, you know, soldering to a PCB, just realized in a different regime. ⁓ All that is to say, the assumptions that one makes really rely on there just not being any of these non-equilibrium quasi particles. And they're kind of a vestigial, know, they're vestigial degrees of freedom that, know, ⁓ If we were working with a more natural quantum system, per se, maybe we wouldn't worry about as much. So the fact that we have to cool down the low temperatures, there are a couple of reasons for that. But maybe first and foremost is to suppress the thermal excitation of quasi particles in our superconducting films. ⁓ And yeah, if you don't get rid of them or you can't shield away from them, it's ⁓ just a very natural set of degrees of freedom for your qubit to couple to and exchange information with. And that leads to errors. So it's very foundational in the way that we choose to encode information in superconducting circuits that there's going to be some limitation at some level. from quasi particles. Even if you cool down really, really, really low and you get the quasi particle population down basically as low as you can, you can end up in a situation where there's just one quasi particle left in the circuit and that can still be a limiting. ⁓
Daniel Rodan Legrain: So let me follow up a little bit on that idea. And how should we think about quasi-particles in terms of, let's say, their direct impact on device performance and scaling?
Kyle Serniak: ⁓ Yeah, quasi-particles are fermionic excitations in the superconducting condensate. They're just some excitations out of the ground state of the BCS, BCS ground state of the superconductor. And whenever they tunnel across a Josephson junction, or really it's whenever they're living in a place that you have current flowing or phase drop in the superconducting circuit. ⁓ they can couple to that phase degree of freedom that we use to describe ⁓ the state of the qubit and take energy. And so ⁓ there's a component of that that's gonna happen in steady state. There's just always can be some number, maybe it's very, very small number of quasi particles living in the qubits ⁓ that could limit sort of on on average, over in steady state, the performance of the qubits. But more recently, it's been identified that ionizing radiation and even mechanical relaxation of the chip itself or mechanical induced excitations from the cryostat into the chip of the superconducting qubit processor ⁓ can create quasi particles and Via that mechanism, you can be coupling in errors from ionizing radiation or mechanical excitations into the qubits through the quasi particles. And those in particular ⁓ are a little bit difficult to deal with because they typically result in errors that are correlated in both space on a processor and in time. So these spatiotemporally correlated errors ⁓ pose some challenges for error correction, specifically in decoding, where a lot of the assumptions that go into those protocols assume just completely random ⁓ errors. So they're definitely a challenge. ⁓ And folks are working on mitigating them. So in classical electronics, ⁓ folks have been able to engineer radiation-hardened classical CMOS electronics. ⁓ And we're just trying to do the same at this point for superconducting qubits to make them resilient against those types of pretty high energy errors.
Daniel Rodan Legrain: All the best in that endeavor, Kyle. Actually, you did much of this ⁓ work on quasi-particles during your own PhD at Yale. And I wanted to ask you, what was it like to work in Michel Devorets group, both scientifically and culturally, ⁓ especially given its recent recognition for the 2025 Nobel Prize in Physics?
Kyle Serniak: Thanks. Mm-hmm. Yeah, no, it was great. You know, the culture at Yale was something that went beyond Michelle's group. There's a really great quantum community there, with Rob Schoelkopf, Steve Gerben, Lena Glasman. Now there are more folks. Shruti Puri is there, Mike Hatchridge, Conrad Leonard. There's a bunch of folks, and I'm sure I'm forgetting some, and I apologize. but, you know, there was always, there was always just a really great culture of communication and collaboration between all of the parties there. And I would say, you know, one of the biggest, one of the biggest things that I'm thankful for, for my PhD is doing research in that type of environment that was so, ⁓ was so, was so broad. Like there was always someone who was an expert in, you know, the thing that you were wondering about. And, you know, there was just an expectation that, you know, you could just walk down the hall and find that person and talk to them, get their insight and pay it forward, like, you know, provide it in return when folks are looking for that information. So that was all maybe, you know, I look back very fondly, ⁓ you know, at the culture and the folks I was working with and the friends that I made there. ⁓ I look back really fondly on that. I mean, it's great, know, at APS, the APS meeting was just, ⁓ was that last week, two weeks ago, ⁓ two weeks ago, yeah, time flies. ⁓ But, know, seeing everybody there and being able to catch up and seeing all the great work that folks are doing, ⁓ it continues to be inspiring from that.
Daniel Rodan Legrain: Two weeks ago.
Kyle Serniak: And then, from the technical side, ⁓ again, there was just this culture of just immense support. I remember spending a lot of time in the clean rooms with Lou G. Fruenzio ⁓ fixing tools. And he set a great example of sort of how to be generous with your time and how to help out your colleagues. ⁓ And so I feel very, yeah, feel very privileged to. I often think about some of the classes I took. I took a mesoscopic physics class that Michelle taught and a micro engineering course that Rob taught. And I sat in on a many body physics ⁓ course that Lien had taught. Learning sort of this foundational material from folks who are really, really, really the experts in the material and then having Luigi be the person in the clean room, showing me the ropes was, I just feel very lucky.
Daniel Rodan Legrain: sounds like a really stimulating environment to be. Let me actually ask you, was there a particular moment during that time of your PhD where, within all these interactions, your understanding of qubit physics, let's say fundamentally changed?
Kyle Serniak: Yeah, I don't know. A million. Honestly, one, this will be a little bit self-deprecating. But I definitely remember a time I was sitting at a computer. I tell this story somewhat often, so I apologize if you've heard this before. Sitting at a computer with Sokko Minov, and he was taking some data and he showed me, was like, oh, look at this, it's really cool. And I'm looking at it and... And it's like, you know, a telegraph signal like, it's kind of some signal is noisy and then it jumps and then it jumps and it jumps. And I started thinking to myself like, oh, you know, that's really interesting. Yeah. Remember something wrong with the setup. I think we were debugging things, you know, over the course of the week. and he I'm like, yeah, maybe it's this, maybe it's this, maybe this. He's like, Kyle, just just assume it's assume everything's working. OK. And I'm thinking to myself and and. you know, after a few minutes, not minutes, know, tens of seconds or whatever, he's like, Kyle, it's just quantum jumps. We're actually measuring a qubit now and everything's actually working the way that you'd expect. And we're able to get signal shot readout and resolve the state of the qubit basically in real time. I was like, ⁓ okay. So maybe this actually is gonna work. Maybe there's a, I think that was the first time it kind of clicked for me that everything that was going on was... was real interesting physics that we could understand.
Daniel Rodan Legrain: physics works!
Kyle Serniak: Occasionally it does, yeah. Similar story about cooling down, cooling down a, I think it was a tie nitride or an ibuprofen nitride film, doing four probe measurements, seeing the locking go all wonky, and Heiss, the postdoc I was working with, saying, ⁓ yeah, it's really a superconductor. Anyways.
Daniel Rodan Legrain: No, I mean, I actually, the superconducting story actually resonates with me the first time I saw it. So, very, very, very interesting story.
Kyle Serniak: Yeah. Yeah, it's really a wonder.
Daniel Rodan Legrain: Maybe one more question on your own work on quasi-particles. Is there any way changed the way in how you think or how the community ⁓ thinks or looks under the hood of a qubit, so to say?
Kyle Serniak: Mm-hmm. Yeah, sure. ⁓ So maybe from a couple of different avenues, and not to take credit, ⁓ some things that maybe it pointed out for some folks. The experiments that we were doing ⁓ were, we weren't the first to do them. So these charge parity measurements where you can map. ⁓ It's called the charge parity, which is the number parity of charge carriers that have tunneled across a Josephson junction in a circuit. You can map that to the state of a qubit. And when a quad-sparkle tunnels across the junction, it changes this charge parity. And so you can look for correlations in the change of the charge parity with change of the state in the qubit. In doing so, can pull out which types of errors that you have that are actually coming from quasi particles versus something else. And this was something that Leo de Carlo's did a few years before we started working on it. it's a set of techniques that I think was... you know, not unique for its time by any means, but ⁓ helped push the field a little bit in the direction of really targeted studies of using sort of the right tools for the, like asking the, you know, specific question. ⁓ Because you can disentangle, you know, that type of effect from a simple coherence measurement or a simple fidelity measurement. There's an additional signature, experimental signature that you can use to really identify when ⁓ when there's an error that is correlated with quasi particles or not. So that's really helpful in identifying ⁓ the dominant errors, ⁓ error channels that were happening at that time. And so with that, we noticed something quite interesting, which I think thankfully now the field generally has a good handle on, ⁓ which is that Not only were quasi particles causing errors in our qubits, but they were causing a lot of qubit excitation in addition to relaxation of the qubit. And that was inconsistent with our picture of quasi particles being relatively well-thermalized, living close to the gap edge above the superconducting gap. And that caused us to think a little bit harder about what was going on there. And... realized that it was ⁓ a phenomenon that was relatively well known from... early 2000s of photon assisted tunneling. So basically what you can have is ⁓ photons with high enough energy can couple into the qubit and create a pair of quasi particles right at the Josephson junction because that's where the voltage drop can happen from these photons. So you kind of couple right at that part of the circuit in least a trans bond. And what that does is it changes the charge parity, but not in a way that's correlated at all with quasi particles already existing in the leads. ⁓ It's just the presence of this infrared radiation that can couple to the junction. And so ⁓ we identified that and then got people thinking about, ⁓ that's interesting. You know, I have a I have a qubit that's operating at 20 millikelvin, but the effective temperature of that qubit is something like 100 millikelvin. ⁓ And it caused folks to really reevaluate the way that they're doing filtering and shielding in their setups, which has really helped people in a variety of ways. ⁓ Certainly, you know, the average qubit is a little bit colder today than it was 10 years ago. ⁓ ⁓ But also it can be a signifier of other issues with filtering and shielding that could be affecting other types of noise and dissipation and decoharass. So. long-winded explanation of maybe some interesting things that came out of that technical understanding, those experiments.
Daniel Rodan Legrain: Fantastic. Indeed, fantastic work there, Kyle. ⁓ Let's go from underlying physics and understanding it to my next question on how we can translate that into designing better qubits. So you in particular at Lincoln Labs work on...
Kyle Serniak: Mm-hmm.
Daniel Rodan Legrain: circuit design, nose mitigation, and new architectures, also with your work at MIT. ⁓ What are the design principles that feel most promising to you for ⁓ improving qubit performance? And what trade-offs, I would say, are often underestimated?
Kyle Serniak: Hmm. So I think there's a lot of optimism right now that, you know, we've identified, we've identified a lot of challenges, but people are jazzed to really come up with creative solutions to them. You know, I would say it's not contentious to say that most of the field really views ⁓ two level system defects, dielectric loss from two level system defects as one of the primary challenges facing superconducting qubits. This is also kind of stems in the same way that I was talking about with the quasi particles. The fact that we're able to construct these circuits out of, I won't say arbitrary materials, but somewhat arbitrary materials as long as they have some properties. They need to be low loss and from a practical standpoint, they need to be superconducting to reduce the number of charge carriers that the qubit can exchange energy with. There's a lot of freedom outside of that, whether you're using aluminum metalization or niobium or tantalum or rhenium or tinitride or niobium tinitride. There's a whole suite of materials that one can use to construct these circuits. the presence of, somewhat ubiquitous presence of chillable system defects to the degree that we understand them in these materials and in the way that we construct the qubits. ⁓ is definitely something that we have to contend with. And folks are getting more more creative about how to understand it. ⁓ Kind in the same way that those charge parity measurements that were developed gave a pretty clear lens into the properties, or not properties, but into the effects of quasi particles on superconducting qubits. ⁓ Folks are getting creative and trying to do similar types of studies that are specifically designed at probing tool system defects. ⁓ And that's really the name of the game for improving superconducting qubits, these relatively engineerable, I mean, it's a word that we use a lot, but they're engineerable in the sense that there are a lot of choices that go into ⁓ their construction. that the experimentalist has a good degree of freedom to vary and modify and choose and optimize. ⁓ So designing specific experiments and there's intersectionality there between the experiment and the device design and then the metrology technique, like what actual experiment measurements are you doing? It's that whole picture holistically that really ⁓ folks are really grabbing onto and recognizing You know, there's a way to solve these problems. There's a way to identify the issue and do comparisons to solve the problems. And it's the recognition that there is this intersectionality, I think, ⁓ that people are really, that's going to really accelerate ⁓ the work in this space over the next couple of years.
Daniel Rodan Legrain: me push a little bit in that direction. as we see systems scaling, where do you think are the main bottlenecks? Are we talking about materials here and device physics? Or maybe are we now talking about system levels ⁓ challenges, packaging control, calibration?
Kyle Serniak: I, you know, it's hard to pull out just one. They're all interrelated. ⁓ There's another aspect that we haven't really talked about, which is the way that, you know, in this theme of engineerability, how you actually encode the information in a superconducting circuit is also something that folks have some degree of control over. And ⁓ it's kind of all of these different things going in parallel. the materials, the information encoding, how you're doing the control, the filtering and shielding of the experimental setup, all of these different requirements are kind of pushing and pulling on each other in parallel. ⁓ And so, it's hard to pick out a specific thing because I'm a believer that the quantum computer that we have one day really may not look... so similar to what we're using right now. ⁓ I think it's going to have a lot of similarities, ⁓ but it really depends on how you're going to try to encode that information to really understand the most critical pieces. And so right now, it's still a research endeavor. So research on all of those vectors is extremely valuable ⁓ in sort of understanding the trade space of what is going to be the best choice long term.
Daniel Rodan Legrain: So Kyle, mean, a big part of enabling all that progress lies in the fabrication and the access to it. And you have actually been very closely involved with the SquiL ⁓ Foundry. ⁓
Kyle Serniak: Mm-hmm.
Daniel Rodan Legrain: Maybe you can tell us a couple of words of what it is and also why is it important to have access to a shared high quality fabrication facility, especially in the quantum ecosystem today.
Kyle Serniak: Yeah, absolutely. So the Squill Foundry is a service that we provide out of our group, the Quantum Enabled Computation Group at Lincoln Lab. ⁓ It's a sponsored research program from Laboratory for Physical Sciences through their ⁓ Qubit Collaboratory. ⁓ And so maybe it was four years ago or so they decided to set up ⁓ three foundry programs, one for superconducting qubits, is us, and then two for quantum dot qubits at HRL and Intel. And the goal of it was to, as you say, ⁓ provide a vehicle for which the community can receive high quality ⁓ quantum devices from established... ⁓ fabrication processes throughout the country. And so our role within it is to provide superconducting qubit devices in service of folks' to them so that they don't have to worry about the fabrication, the packaging, all of that. And so this has been something, as I said, it's been going for about four years. There's upwards of 50 users now. ⁓ And the way that typically looks is you submit a project application and as long as your project fits within some criteria that the sponsors want us to provide these devices to you, ⁓ you get a design rule checker ⁓ software package to make sure that the designs that you're generating for your experiment are going to, basically are they gonna yield? Are they gonna be able to be fabricated in our process? And you then submit those GDS designs to us. We aggregate them into a multi-user mask and our team fabricates them and then ships them out to the community. So, well, to the person who submitted them, I should clarify. But yeah, so, you know, it's now it's about 50 groups. Even early on folks were able to give presentations and write papers based on these devices and that number has grown. I don't have that number off the top of my head, but it's in the tens of papers that have been published based on devices fabricated through the Squirrel Foundry. ⁓ And that's also not including ⁓ any of the internal work from Lincoln Lab or the work from MIT campus ⁓ that has been using the same fabrication processes for many, many years in service of those experiments. yeah, so people use it for a variety of different ⁓ application models. ⁓ There are some groups who, you know, if you're a new professor starting out, and it's going to take you six months or a year or more to sort of get new fabrication equipment into your lab and installed and develop processes and be able to make, you know, reliable superconducting quantum devices. Submitting to the Squill Foundry is a way to really accelerate that onboarding process of starting up a new lab, for example. We also offer some advanced capabilities. So there are other folks who maybe even have a well-established process for making quality superconducting qubits, but maybe they never took the time to develop airbrush crossovers or flip chip capabilities. And so there's like another layer of complexity that we offer that can also fill a position within within an established research group's operating model. Additionally, tools go down, right? So having a parallel pathway to have devices fabricated is generally useful for any group. If... ⁓ you know, your eBimuthography system is getting a little old or something, you know. ⁓ If you're not confident in its reliability, ⁓ diversifying, ⁓ you know, the ways in which you are acquiring devices and fabricating devices can be really helpful for risk mitigation in those kind of ⁓ schemes. So the SQL Foundry and generally speaking, it's called the Qubits for Computing Foundries. ⁓ ⁓ with LPS, they do a lot of different things for lot of different people. ⁓ But there's really no doubt that it's making a big impact in the community and it's something that our team is again, really thankful to be able to be a part of and be able to support folks that way.
Daniel Rodan Legrain: I mean it is truly a fantastic program and 50 users that grew fast.
Kyle Serniak: Yeah, yeah, I mean there was a there was a initial like seedling year. It would be a year and a half when it started with maybe about 10 users and ever since then it's grown relatively steadily. We onboard new users approximately with every fab run that we do of which there are seven, eight a year that are for the Foundry users. And yeah, basically every, you know, we're constantly onboarding new folks with new.
Daniel Rodan Legrain: More broadly, how do you see Lincoln Labs' ⁓ in the quantum ecosystem compared to universities and ⁓ startups? Especially, I'm thinking in terms of enabling infrastructure, testing and evaluation, and supporting the broader community.
Kyle Serniak: Yeah, I mean, that's kind of singular to the role of any FFRDC in its respective community. so Lincoln Laboratory, I should say, is a much larger institution than just superconducting qubit research. But yeah, central to that mission is... Supporting the community and supporting the sponsor community in in all the ways that that makes sense. That's something that that varies field to ⁓ Typically, ⁓ you know, we're focused on advanced prototyping which which is an interpretable, you know ⁓ You know thing in and of itself ⁓ But you know, it it looks like academic style research done in a professional setting it looks like ⁓ a test and evaluation where our team is working alongside the sponsors to provide feedback on research. Another role that we have within the community is what's called government furnished capability. And for certain programs, the sponsors can, ⁓ you know, fund Lincoln Lab to support just very directly the performers on a specific program. And that can look like a bunch of different things as well. That can look like academic research, that can look like providing devices specifically for the goals of that program, ⁓ providing any expertise. It's an interesting role ⁓ that really relies on the unique relationship that FFRDCs have with the sponsored community. and is generally just a very nice and impactful way to engage with the community that isn't quite natural to do at an academic group or an industry.
Daniel Rodan Legrain: How do you balance standardization with the need for innovation here?
Kyle Serniak: That's a great question. you can't, you know, there's no point in standardizing without innovation, right? ⁓ If you standardize and you're really good at one thing and the next good thing comes along, you have to be ready to incorporate that. Otherwise, the work that you're doing isn't quite as relevant anymore. So we have really healthy efforts on both sides of that line. We really want to make sure that we, at any given time, can make really high quality qubits ⁓ for experiments that maybe, you know, way to describe them is anything that isn't fabrication development focused. Like we want to have something that we know is just going to work when we're designing a new qubit circuit or trying to develop a new metrology protocol or... doing something that's focused on the rest of the infrastructure, the measurement chain, the way we're controlling the qubits. So we wanna make sure that there's always a complete process flow, fabrication process flow that is extremely well vetted, extremely reliable, and that we know it's just gonna work. And then on top of that, once that foundation is there, We spend a lot of time doing materials R &D, processing R &D, trying to understand what is the next change that we can make to get a performance improvement on one axis or another. Maybe it's cubic coherence. ⁓ Maybe it's the reliability of the Josephson Junction critical current targeting. And within our group, This is something that's very central to our sort of identity is we have the full matrix in our minds of, is this going to affect this other axis? And do we need to worry about that? And we're constantly doing the vetting and validation to ensure that this one thing that maybe is great for one metric, is it going to come back to cause a problem? you know along some other axis. So yeah. No, no I'm rambling a bit I was just gonna say ⁓ that like to your original question you know you can't be stagnant you always have to be developing.
Daniel Rodan Legrain: Let me ask you one for... ⁓ sorry, Kyle. So let me ask you a looking forward question here, following up on what you said. And where do you see the biggest opportunities today to keep improving superconducting qubit performance?
Kyle Serniak: Yeah, I mean, I think it's you know, similar to what I was saying before There are all of these different avenues that one can pursue to improve qubit performance a You know a pretty you know, pretty exciting one is the materials Because basically as you improve the materials that we're using for superconducting qubits Basically everything else gets lifted by that ⁓ By that found it's it's kind of the foundation everything gets lifted by by those improvements in a similar way ⁓ The same can also be said for the you know novel qubit circuits where the information is encoded in different ways You can kind of just flip the logic and say ⁓ well for any given Material system if you can if you can encode information in a way that's less sensitive to the defects in those materials Well, everything is just going to get better even with the existing you know, construction methods and materials. The same can be true of if you just improve those, then everything just gets better. So the materials, would say, are really a big... There's a lot of opportunities for improvement and a lot of things that we know we're learning more and more what to avoid, how to avoid it.
Daniel Rodan Legrain: Let me switch gears a little bit, Kyle, before we go to the ⁓ question from the audience. And let's talk about the people that are building those systems. ⁓ And from your point of view, what capabilities or mindsets do you think are becoming essential for the next generation of ⁓ quantum hardware builders, let's say beyond traditional physics training here?
Kyle Serniak: Yeah. So, I mean, very much central to the notion that there are all of these different, you know, there are a lot of things that need to go right ⁓ when making supernetic quantum processors. And I think it's natural maturity of a field to where, you know, the knowledge that was maybe 15 years ago, condensed into everyone's head of a little bit of cryogenics, a little bit of micro-engineering, a little bit of fabrication engineering, a little bit of this, a little bit of that. It's very reasonable for it to be more distributed and that allows for much, much more depth of understanding of all those individual components. And so... For folks who are interested in getting into the field, there's still a lot of value in the general superconducting qubit educational sort of pathway, but there's a lot of space for people who aren't yet in the field. Folks who have done process engineering at... at industrial foundries for, you know, and built up experience that way. That gives, that gives a completely different perspective than someone coming with a PhD from super 90 qubits. ⁓ Similarly, know, professional micro engineers are amazing, right? They, they can do, they can, they, understand the problem and, and, know how to implement solutions just much more effectively than, than than I do ⁓ as a armchair micro-engineer maybe. ⁓ So whether it's cryogenics engineering, thermal management, micro-engineering, fabrication process development, ⁓ quantum algorithm, the field is becoming a little bit more specialized, but in a way that is good and natural for... ⁓ you know, for a field that's and immature.
Daniel Rodan Legrain: What advice would you give to people choosing between, let's say, academia, research labs and startups?
Kyle Serniak: I think, you know, learn as much as you can about the real differences ⁓ because in the field, ⁓ you know, everybody has a role to play and there's a lot of overlap, but, you know, thinking about it day to day, if you look at the situation that you're in, you know, the position that you're in and you see, ⁓ I really, really like this or I really, really don't like this. You know, it sounds a little bit silly, you know, do the due diligence to figure out where you can maintain the things that you really like and where you maybe don't have to do the things that you don't like. If you don't like teaching, maybe academia is not the right spot. If you really love research and nothing else, maybe all of those could be interesting options for you. ⁓ And it's more about the secondary things.
Daniel Rodan Legrain: So one last question from Mikhail. What do you wish you had been told as a first year PhD student?
Kyle Serniak: ⁓ well, that's a good question. So, so many things. I, I wish I had been told, this is actually something I've noticed a lot. So, so moving into a PhD, ⁓ and especially kind of the, middle years of a PhD, is the first time for some folks, like if you, if you haven't worked in between, like if you go straight through school.
Daniel Rodan Legrain: you
Kyle Serniak: ⁓ It can be the first time that you are in a situation where you're doing work and there's no longer sort of the six month or four month for a semester timeline associated with a project. ⁓ And that's something that people often get really bogged down with. The notion of open-ended research being something that can take years and years when You're used to going through school and taking a class and having a final exam or final project and everything being really well contained in this four month ish timeline. And as soon as as soon as you're out of that and really deep in the research, that's that's something that you you can structure in a way that looks similar, but it's pretty easy to to sort of, ⁓ you know, not have that structure. So getting used to getting used to the fact that research can be relatively long-term endeavor. ⁓ I'm not sure the right advice that I wish I had gotten in that respect, ⁓ but I think that's something that folks consistently struggle with, kind of in the middle of a PhD ⁓ that I've certainly noticed and felt myself.
Daniel Rodan Legrain: Thank you, Kyle, for this and all the other insights you shared with us today. ⁓ We have a few minutes for questions from the audience. I see quite a few here populating the chat. ⁓ I see quite a few questions about the squirrel.
Kyle Serniak: Of course. Yeah, no, thank you, Danny.
Daniel Rodan Legrain: program. Let me group them together into one. You know, if I'm a new PI, how can I collaborate with Lincoln? Do you take industry partners? How long does it take to fabricate and ship these devices? What do you have to say,
Kyle Serniak: Yeah. Sure. Yeah. Yeah, so if you are interested in working with us, so if you just Google Squill Foundry, ⁓ there should be a webpage that comes up with an email address that you can send. I believe it's squillfoundry.ll.mit.edu. And from that, you can get all the information that you need about becoming a part of that ecosystem. ⁓ We do have engagement models for all types of institutions, including industry. So don't let that scare you away, ⁓ wrote about that. yeah, in terms of collaboration, absolutely, it's something that we do. ⁓ It depends a bit on the circumstances and it needs to be something that makes sense for us to work more closely together on, but that's also something that we certainly do. ⁓ I don't remember, I think there was going be one more question in there, but...
Daniel Rodan Legrain: That was how long does it take to ship a device.
Kyle Serniak: ⁓ yeah, good question. So it depends a bit. It depends on, you know, whether tools are working as expected or not. Typically, ⁓ it's about eight weeks, eight to 10 weeks ⁓ for between submission of the designs and shipping of the devices. ⁓ Sometimes it's shorter, sometimes it's longer.
Daniel Rodan Legrain: I mean it's really amazing that you design a chip and then it comes to your mail without needing a whole cleanroom in your university.
Kyle Serniak: Yeah, it's, yeah, and that's, I'll say that is something that, you know, the timeline is longer than what a lot of academic groups can do themselves. And, you know, it's just a different way of working. It's not long, as I understand from a commercial foundry perspective. And... You know, it just, it changes the way you think about an experiment. You can't do as much serial development, ⁓ but you can really parallelize the things that you're doing and really answer questions that way. Different way of thinking, but similarly effective.
Daniel Rodan Legrain: Thank you, Kyle. ⁓ We have a student here who doesn't have quantum courses in his university and is asking, how can we learn about all you're sharing with us today without necessarily having to go to a classroom that I don't have access to?
Kyle Serniak: That's a great question. So the. I'm now forgetting. I mean, there's a lot of online resources specific to superconducting qubits or quantum information in general. I remember, you know, the IBM Qiskit materials being really, really helpful for folks onboarding into the field. ⁓ And there are a variety of other programs. Qubit by Qubit is one that comes to mind that's focused, I think, on a summer course on... ⁓ on, you know, that focuses on supergnetting qubits as well as more quantum information generally. But, you know, honestly, there's a network of folks. There's a network of folks, and if you reach out to them, then, you know, we have a pretty friendly community, and any connection that you have, will likely be happy to share with whoever they're... their favorite resources, the things that they find most valuable.
Daniel Rodan Legrain: I agree. Let me also mention that this Fireshot Chat series Quantum Builders is a good starting point to get to know who is available out there.
Kyle Serniak: Yeah, there you go. That's right. That's right.
Daniel Rodan Legrain: And Kyle, we have time for one more question. And you mentioned something that the quantum computer of the future might be different to the ones we have today. And I guess someone picked on that nuance and would be curious to hear yourself on how does the quantum computer of the future would look like.
Kyle Serniak: I don't, mean, so generally speaking, ⁓ I mean, it could, it could look like what we have today. I think people have, people have generally, you know, a lot of the optimism in the field comes from the fact that what we have is pretty good. And it's, the academic in me and the academic and a lot of people who, ⁓ who, who really think of themselves, yeah, there must just be, surely there's something a little bit better. And The fact that, especially for super-uniting quantum processors, the way you do control, the way you encode information, the materials that you're using, all of these things are really up to the experimenter, the designer, to really optimize and really understand and balance different trade-offs. I mean it more as a general statement that I'd be surprised if it was so similar to what we have today. ⁓ I think it's all of those avenues that there's a lot of room for improvement.
Daniel Rodan Legrain: Time will tell.
Kyle Serniak: Time will tell.
Daniel Rodan Legrain: Kyle, thank you so, so much for joining us today. I mean, your work really highlights something. Actually, we also really strongly believe at QBlocks that progress in quantum computing comes from understanding the underlying physics, building strong infrastructure, and enabling collaboration across the ecosystem. ⁓ As a small token of appreciation, we've sent you one of our QBlocks mugs with our motto, building quantum together. Hopefully a good companion for your next round of experiments. ⁓ For everyone watching, the recording will be available shortly on the QBlocks YouTube channel, on Spotify and Apple Podcasts under Quantum Builders. And please stay tuned for future Farsight chats. We have many more exciting gets ahead. And Kyle, thank you again. Thank you for joining us. Thank you for the amazing...
Kyle Serniak: Mm-hmm.
Daniel Rodan Legrain: insights and thank you for everyone joining us today.
Kyle Serniak: Yeah, thank you, Danny. This was a lot of fun. It was really great chatting with you and thanks to everybody for tuning in.
Kyle Serniak: Hi, Danny.
Daniel Rodan Legrain: I'm Daniel Rodan Legrain. I'm the Quantum Science Engineering Lead for North America at Qblogs, the sponsor of today's And these five chat series basically brings together series shaping the quantum ecosystems ⁓ all the way across academia, research institutions, and industry. And in previous editions, we've had the pleasure of hosting Will Oliver, Berta Jong, Peter Shore, Steve Girvin, Gretchen Campbell, Chuck Black, David Afshalom, among others across the stack. And today, I'm very happy and excited to welcome Dr. Kyle Cernyak. ⁓ Kyle, you're the assistant group leader in the quantum enabled competition group at the MIT Lincoln Lab, your research affiliate at ⁓ MIT RLE and ECOS, and part of the MIT Center for Quantum Engineering. ⁓ Kyle's work focuses on improving superconducting quantum devices, basically spanning noise, quasi-particles, materials, device design, fabrication, and also on building shared infrastructure for the community, including the Squill Foundry. Kyle, you completed your PhD in Michel de Voré's group at Yale, working on non-equilibrium quasi-particles, and a topic that I would say remains very central to qubit performance today. ⁓ Kyle, again, it's a real pleasure to have you today here. it.
Kyle Serniak: Thanks for the introduction Danny. Yeah, I'm really excited to chat with you today Thank you so much for inviting me to do this and I'm looking forward to sharing Yeah, yeah sharing this discussion with you
Daniel Rodan Legrain: Fantastic. Let's dive in. Before that, for everyone ⁓ watching live, please feel free to post your questions in the chat, and I will bring a few at the end of the discussion. So, Kyle, why don't you start with your own journey? I mean, how you ended up basically working at the intersection of device physics, fabrications, and community infrastructure. So...
Kyle Serniak: Yeah, yeah, happy to. So I started out in this field as an undergraduate. I was completing my undergraduate degree at Florida State University. one of the, you know, a really great resource that exists there is the National Hamadetic Field Laboratory. And I was a physics major and, you wanting to get into research. I had already done some research in the chemistry department, but wanted to make a transition more into physics. With the magnet lab there, I was looking for opportunities sort of therein and got involved with Irinal Kirescu, who was one of the early folks working on control of flux qubits. was a postdoc in Delft doing that. With Hans Moi. And ⁓ yeah, got involved there. And that was kind of where I started getting into microbe engineering and all of the sort of underpinnings of the superconducting qubit field. And was also where I first experienced sort of the big lab kind of atmosphere. ⁓ of the National Ammonautical Field Lab at the, well, it's today, but there in Tallahassee, which kind of pointed me toward my trajectory also in super90 cubits at relatively large research institutions, MIT Lincoln Lab.
Daniel Rodan Legrain: Wonderful. Was there any key moment that made you realize that this is the field you wanted to be focusing on?
Kyle Serniak: I think it was, I think it was just a confluence of things. You know, there was opportunity specifically, you know, to do research related to this field at the Mag Lab and, and, you know, without, without that, I don't know that I would have ended up on this trajectory, to be honest with you. So definitely, you know, a target of opportunity and... And I think I knew that I would stick with it once I got in the lab. I saw all the different aspects that kind of contribute to the field. The fabrication side, the RF design side, the cryogenic side. It felt very dynamic. It felt very exciting. It felt like I would never feel like I'm doing the same thing day after day after day, which is certainly true, I would say, even now. ⁓ I think it was the combination of those things that made it click for me that ⁓ I had already known pretty strongly that I was interested in research at that time, ⁓ but really that this field was the right fit for me was probably around then.
Daniel Rodan Legrain: And here we are now, at the right place at the right time.
Kyle Serniak: You know, ⁓ it's easy to say that everything is merit-based and if you work hard, you get there, but there's definitely an aspect of luck, I would say.
Daniel Rodan Legrain: So talking about that, I mean, you now operate across MIT and Lincoln Lab. How would you actually describe kind of like the value of that model in practice? And what does that environment enable you to do that maybe would be difficult to achieve in just one of the two settings?
Kyle Serniak: Yeah, no, it's a question. maybe for a little bit of context, ⁓ MIT Lincoln Laboratory is a federally funded research and development center that's operated by MIT. So everybody who's a researcher out at Lincoln Lab, we're all MIT employees. And there's been a long history of superconducting qubit work. ⁓ both at MIT starting with Terry Orlando and then at Lincoln Lab with Will Oliver when he was here, along with Mark Gouker, a bunch of other folks that kind of started those efforts 25 years ago at this point, thereabouts. And so there's a lot of history there and there's always been a history of tight coordination and collaboration between the teams out at Lincoln Lab and the teams at MIT campus. So, you know. It's a really great environment to work in. On the Lincoln Lab side, have staff scientists, generally speaking, with a lot of institutional knowledge, people who love what they're doing and have been doing it for many, years. ⁓ And from the campus side, it's an academic group. So you've got students and post-docs with fresh ideas and, you know, just boundless enthusiasm coming through and bringing those folks together, ⁓ you know, it can make for some really exciting research and development. ⁓ So, you know, I love it, to be honest with you. ⁓ It's a great situation to work within. I'm thankful every day that I'm able to contribute to this ⁓ research ecosystem.
Daniel Rodan Legrain: I can say I was fortunate enough to leave it during my own career and I can say it's a really special place and you are definitely one of the key pieces to make it happen and work, Kyle.
Kyle Serniak: Yeah, yeah, Yeah, well, thank you. Thank you for saying that. guess we can break the fourth wall a little bit and say that, that we worked a little bit together when you were a postdoc in the Engineering and Quantum Systems group, Wallerbergs group at MIT, ⁓ working on 2D materials and gate mod qubits and things like that. It was a lot of fun.
Daniel Rodan Legrain: Indeed. ⁓ I wanted to also ask you, Kyle, was there anything, something that maybe surprised you a little bit more when you moved to a, let's say a more engineering driven environment?
Kyle Serniak: being like a laboratory.
Daniel Rodan Legrain: Sorry, Bing.
Kyle Serniak: Like Lincoln Laboratory being the more engineering driven environment. Yeah, there were some adjustments, ⁓ but really, you know, maybe less so from an engineering perspective and more a, ⁓ you know, professional perspective. So Lincoln Lab operates in an interesting space between academia and industry. ⁓
Daniel Rodan Legrain: Indeed.
Kyle Serniak: the way the day-to-day looks is maybe more similar to what you'd expect in industry. ⁓ It's a little bit more structured, little bit more, ⁓ yeah, structured is the right word. ⁓ But we're doing a lot of academic style research. ⁓ so getting used to doing really the same things that I did during my PhD, designing. fabricating, measuring, analyzing data, doing experiments ⁓ in a different setting that was a little bit more methodical is maybe the right way to put it. In a PhD program, you're kind of go, go, go, and you're kind of doing everything yourself. ⁓ In a more professional setting, there's more distribution of responsibilities that makes for really, really excellent work and really methodical work. that's just a bit of a shock, you know, moving from one to the other. And there's pros and cons to both, but I definitely really enjoy the more methodical, the more. which is forward looking, let's say, ⁓ in the way that we structure the work at it.
Daniel Rodan Legrain: Thank you, Kyle. Let's maybe now dive into the physics and maybe zoom into one of the central themes of your work, what actually limits ⁓ qubits performances. So a lot of the field talks about T1, T2, but your work actually goes much deeper than that into the underlying mechanisms, especially non-equilibrium quasi-particles. What makes them such a persistent limitation and why do they still matter so much today?
Kyle Serniak: So ⁓ quasi particles, non-equilibrium quasi particles are interesting because they're just so tightly ingrained with a lot of the interesting and kind of weird parts of superconducting qubits as a hardware modality. ⁓ We often talk about superconducting qubits as artificial atoms, right? There's this picture that they have complex. ⁓ energy level structure and interesting selection rules sometimes between, you know, for transitions between states. Just like, ⁓ you know, a trapped ion or one of these more archetypically quantum kind of systems. ⁓ But the only reason that we can ever really resolve any of that quantum phenomena is if we're in the limit where we can just completely forget about all of the electronic degrees of freedom in these circuits. You know, we write down a superconducting qubit with, you know, inductors and capacitors and okay, there are Josephson junctions, which are a little bit, a little bit non-standard from a circuit perspective. But otherwise we're talking about components that people think about holding in their hand and, you know, soldering to a PCB, just realized in a different regime. ⁓ All that is to say, the assumptions that one makes really rely on there just not being any of these non-equilibrium quasi particles. And they're kind of a vestigial, know, they're vestigial degrees of freedom that, know, ⁓ If we were working with a more natural quantum system, per se, maybe we wouldn't worry about as much. So the fact that we have to cool down the low temperatures, there are a couple of reasons for that. But maybe first and foremost is to suppress the thermal excitation of quasi particles in our superconducting films. ⁓ And yeah, if you don't get rid of them or you can't shield away from them, it's ⁓ just a very natural set of degrees of freedom for your qubit to couple to and exchange information with. And that leads to errors. So it's very foundational in the way that we choose to encode information in superconducting circuits that there's going to be some limitation at some level. from quasi particles. Even if you cool down really, really, really low and you get the quasi particle population down basically as low as you can, you can end up in a situation where there's just one quasi particle left in the circuit and that can still be a limiting. ⁓
Daniel Rodan Legrain: So let me follow up a little bit on that idea. And how should we think about quasi-particles in terms of, let's say, their direct impact on device performance and scaling?
Kyle Serniak: ⁓ Yeah, quasi-particles are fermionic excitations in the superconducting condensate. They're just some excitations out of the ground state of the BCS, BCS ground state of the superconductor. And whenever they tunnel across a Josephson junction, or really it's whenever they're living in a place that you have current flowing or phase drop in the superconducting circuit. ⁓ they can couple to that phase degree of freedom that we use to describe ⁓ the state of the qubit and take energy. And so ⁓ there's a component of that that's gonna happen in steady state. There's just always can be some number, maybe it's very, very small number of quasi particles living in the qubits ⁓ that could limit sort of on on average, over in steady state, the performance of the qubits. But more recently, it's been identified that ionizing radiation and even mechanical relaxation of the chip itself or mechanical induced excitations from the cryostat into the chip of the superconducting qubit processor ⁓ can create quasi particles and Via that mechanism, you can be coupling in errors from ionizing radiation or mechanical excitations into the qubits through the quasi particles. And those in particular ⁓ are a little bit difficult to deal with because they typically result in errors that are correlated in both space on a processor and in time. So these spatiotemporally correlated errors ⁓ pose some challenges for error correction, specifically in decoding, where a lot of the assumptions that go into those protocols assume just completely random ⁓ errors. So they're definitely a challenge. ⁓ And folks are working on mitigating them. So in classical electronics, ⁓ folks have been able to engineer radiation-hardened classical CMOS electronics. ⁓ And we're just trying to do the same at this point for superconducting qubits to make them resilient against those types of pretty high energy errors.
Daniel Rodan Legrain: All the best in that endeavor, Kyle. Actually, you did much of this ⁓ work on quasi-particles during your own PhD at Yale. And I wanted to ask you, what was it like to work in Michel Devorets group, both scientifically and culturally, ⁓ especially given its recent recognition for the 2025 Nobel Prize in Physics?
Kyle Serniak: Thanks. Mm-hmm. Yeah, no, it was great. You know, the culture at Yale was something that went beyond Michelle's group. There's a really great quantum community there, with Rob Schoelkopf, Steve Gerben, Lena Glasman. Now there are more folks. Shruti Puri is there, Mike Hatchridge, Conrad Leonard. There's a bunch of folks, and I'm sure I'm forgetting some, and I apologize. but, you know, there was always, there was always just a really great culture of communication and collaboration between all of the parties there. And I would say, you know, one of the biggest, one of the biggest things that I'm thankful for, for my PhD is doing research in that type of environment that was so, ⁓ was so, was so broad. Like there was always someone who was an expert in, you know, the thing that you were wondering about. And, you know, there was just an expectation that, you know, you could just walk down the hall and find that person and talk to them, get their insight and pay it forward, like, you know, provide it in return when folks are looking for that information. So that was all maybe, you know, I look back very fondly, ⁓ you know, at the culture and the folks I was working with and the friends that I made there. ⁓ I look back really fondly on that. I mean, it's great, know, at APS, the APS meeting was just, ⁓ was that last week, two weeks ago, ⁓ two weeks ago, yeah, time flies. ⁓ But, know, seeing everybody there and being able to catch up and seeing all the great work that folks are doing, ⁓ it continues to be inspiring from that.
Daniel Rodan Legrain: Two weeks ago.
Kyle Serniak: And then, from the technical side, ⁓ again, there was just this culture of just immense support. I remember spending a lot of time in the clean rooms with Lou G. Fruenzio ⁓ fixing tools. And he set a great example of sort of how to be generous with your time and how to help out your colleagues. ⁓ And so I feel very, yeah, feel very privileged to. I often think about some of the classes I took. I took a mesoscopic physics class that Michelle taught and a micro engineering course that Rob taught. And I sat in on a many body physics ⁓ course that Lien had taught. Learning sort of this foundational material from folks who are really, really, really the experts in the material and then having Luigi be the person in the clean room, showing me the ropes was, I just feel very lucky.
Daniel Rodan Legrain: sounds like a really stimulating environment to be. Let me actually ask you, was there a particular moment during that time of your PhD where, within all these interactions, your understanding of qubit physics, let's say fundamentally changed?
Kyle Serniak: Yeah, I don't know. A million. Honestly, one, this will be a little bit self-deprecating. But I definitely remember a time I was sitting at a computer. I tell this story somewhat often, so I apologize if you've heard this before. Sitting at a computer with Sokko Minov, and he was taking some data and he showed me, was like, oh, look at this, it's really cool. And I'm looking at it and... And it's like, you know, a telegraph signal like, it's kind of some signal is noisy and then it jumps and then it jumps and it jumps. And I started thinking to myself like, oh, you know, that's really interesting. Yeah. Remember something wrong with the setup. I think we were debugging things, you know, over the course of the week. and he I'm like, yeah, maybe it's this, maybe it's this, maybe this. He's like, Kyle, just just assume it's assume everything's working. OK. And I'm thinking to myself and and. you know, after a few minutes, not minutes, know, tens of seconds or whatever, he's like, Kyle, it's just quantum jumps. We're actually measuring a qubit now and everything's actually working the way that you'd expect. And we're able to get signal shot readout and resolve the state of the qubit basically in real time. I was like, ⁓ okay. So maybe this actually is gonna work. Maybe there's a, I think that was the first time it kind of clicked for me that everything that was going on was... was real interesting physics that we could understand.
Daniel Rodan Legrain: physics works!
Kyle Serniak: Occasionally it does, yeah. Similar story about cooling down, cooling down a, I think it was a tie nitride or an ibuprofen nitride film, doing four probe measurements, seeing the locking go all wonky, and Heiss, the postdoc I was working with, saying, ⁓ yeah, it's really a superconductor. Anyways.
Daniel Rodan Legrain: No, I mean, I actually, the superconducting story actually resonates with me the first time I saw it. So, very, very, very interesting story.
Kyle Serniak: Yeah. Yeah, it's really a wonder.
Daniel Rodan Legrain: Maybe one more question on your own work on quasi-particles. Is there any way changed the way in how you think or how the community ⁓ thinks or looks under the hood of a qubit, so to say?
Kyle Serniak: Mm-hmm. Yeah, sure. ⁓ So maybe from a couple of different avenues, and not to take credit, ⁓ some things that maybe it pointed out for some folks. The experiments that we were doing ⁓ were, we weren't the first to do them. So these charge parity measurements where you can map. ⁓ It's called the charge parity, which is the number parity of charge carriers that have tunneled across a Josephson junction in a circuit. You can map that to the state of a qubit. And when a quad-sparkle tunnels across the junction, it changes this charge parity. And so you can look for correlations in the change of the charge parity with change of the state in the qubit. In doing so, can pull out which types of errors that you have that are actually coming from quasi particles versus something else. And this was something that Leo de Carlo's did a few years before we started working on it. it's a set of techniques that I think was... you know, not unique for its time by any means, but ⁓ helped push the field a little bit in the direction of really targeted studies of using sort of the right tools for the, like asking the, you know, specific question. ⁓ Because you can disentangle, you know, that type of effect from a simple coherence measurement or a simple fidelity measurement. There's an additional signature, experimental signature that you can use to really identify when ⁓ when there's an error that is correlated with quasi particles or not. So that's really helpful in identifying ⁓ the dominant errors, ⁓ error channels that were happening at that time. And so with that, we noticed something quite interesting, which I think thankfully now the field generally has a good handle on, ⁓ which is that Not only were quasi particles causing errors in our qubits, but they were causing a lot of qubit excitation in addition to relaxation of the qubit. And that was inconsistent with our picture of quasi particles being relatively well-thermalized, living close to the gap edge above the superconducting gap. And that caused us to think a little bit harder about what was going on there. And... realized that it was ⁓ a phenomenon that was relatively well known from... early 2000s of photon assisted tunneling. So basically what you can have is ⁓ photons with high enough energy can couple into the qubit and create a pair of quasi particles right at the Josephson junction because that's where the voltage drop can happen from these photons. So you kind of couple right at that part of the circuit in least a trans bond. And what that does is it changes the charge parity, but not in a way that's correlated at all with quasi particles already existing in the leads. ⁓ It's just the presence of this infrared radiation that can couple to the junction. And so ⁓ we identified that and then got people thinking about, ⁓ that's interesting. You know, I have a I have a qubit that's operating at 20 millikelvin, but the effective temperature of that qubit is something like 100 millikelvin. ⁓ And it caused folks to really reevaluate the way that they're doing filtering and shielding in their setups, which has really helped people in a variety of ways. ⁓ Certainly, you know, the average qubit is a little bit colder today than it was 10 years ago. ⁓ ⁓ But also it can be a signifier of other issues with filtering and shielding that could be affecting other types of noise and dissipation and decoharass. So. long-winded explanation of maybe some interesting things that came out of that technical understanding, those experiments.
Daniel Rodan Legrain: Fantastic. Indeed, fantastic work there, Kyle. ⁓ Let's go from underlying physics and understanding it to my next question on how we can translate that into designing better qubits. So you in particular at Lincoln Labs work on...
Kyle Serniak: Mm-hmm.
Daniel Rodan Legrain: circuit design, nose mitigation, and new architectures, also with your work at MIT. ⁓ What are the design principles that feel most promising to you for ⁓ improving qubit performance? And what trade-offs, I would say, are often underestimated?
Kyle Serniak: Hmm. So I think there's a lot of optimism right now that, you know, we've identified, we've identified a lot of challenges, but people are jazzed to really come up with creative solutions to them. You know, I would say it's not contentious to say that most of the field really views ⁓ two level system defects, dielectric loss from two level system defects as one of the primary challenges facing superconducting qubits. This is also kind of stems in the same way that I was talking about with the quasi particles. The fact that we're able to construct these circuits out of, I won't say arbitrary materials, but somewhat arbitrary materials as long as they have some properties. They need to be low loss and from a practical standpoint, they need to be superconducting to reduce the number of charge carriers that the qubit can exchange energy with. There's a lot of freedom outside of that, whether you're using aluminum metalization or niobium or tantalum or rhenium or tinitride or niobium tinitride. There's a whole suite of materials that one can use to construct these circuits. the presence of, somewhat ubiquitous presence of chillable system defects to the degree that we understand them in these materials and in the way that we construct the qubits. ⁓ is definitely something that we have to contend with. And folks are getting more more creative about how to understand it. ⁓ Kind in the same way that those charge parity measurements that were developed gave a pretty clear lens into the properties, or not properties, but into the effects of quasi particles on superconducting qubits. ⁓ Folks are getting creative and trying to do similar types of studies that are specifically designed at probing tool system defects. ⁓ And that's really the name of the game for improving superconducting qubits, these relatively engineerable, I mean, it's a word that we use a lot, but they're engineerable in the sense that there are a lot of choices that go into ⁓ their construction. that the experimentalist has a good degree of freedom to vary and modify and choose and optimize. ⁓ So designing specific experiments and there's intersectionality there between the experiment and the device design and then the metrology technique, like what actual experiment measurements are you doing? It's that whole picture holistically that really ⁓ folks are really grabbing onto and recognizing You know, there's a way to solve these problems. There's a way to identify the issue and do comparisons to solve the problems. And it's the recognition that there is this intersectionality, I think, ⁓ that people are really, that's going to really accelerate ⁓ the work in this space over the next couple of years.
Daniel Rodan Legrain: me push a little bit in that direction. as we see systems scaling, where do you think are the main bottlenecks? Are we talking about materials here and device physics? Or maybe are we now talking about system levels ⁓ challenges, packaging control, calibration?
Kyle Serniak: I, you know, it's hard to pull out just one. They're all interrelated. ⁓ There's another aspect that we haven't really talked about, which is the way that, you know, in this theme of engineerability, how you actually encode the information in a superconducting circuit is also something that folks have some degree of control over. And ⁓ it's kind of all of these different things going in parallel. the materials, the information encoding, how you're doing the control, the filtering and shielding of the experimental setup, all of these different requirements are kind of pushing and pulling on each other in parallel. ⁓ And so, it's hard to pick out a specific thing because I'm a believer that the quantum computer that we have one day really may not look... so similar to what we're using right now. ⁓ I think it's going to have a lot of similarities, ⁓ but it really depends on how you're going to try to encode that information to really understand the most critical pieces. And so right now, it's still a research endeavor. So research on all of those vectors is extremely valuable ⁓ in sort of understanding the trade space of what is going to be the best choice long term.
Daniel Rodan Legrain: So Kyle, mean, a big part of enabling all that progress lies in the fabrication and the access to it. And you have actually been very closely involved with the SquiL ⁓ Foundry. ⁓
Kyle Serniak: Mm-hmm.
Daniel Rodan Legrain: Maybe you can tell us a couple of words of what it is and also why is it important to have access to a shared high quality fabrication facility, especially in the quantum ecosystem today.
Kyle Serniak: Yeah, absolutely. So the Squill Foundry is a service that we provide out of our group, the Quantum Enabled Computation Group at Lincoln Lab. ⁓ It's a sponsored research program from Laboratory for Physical Sciences through their ⁓ Qubit Collaboratory. ⁓ And so maybe it was four years ago or so they decided to set up ⁓ three foundry programs, one for superconducting qubits, is us, and then two for quantum dot qubits at HRL and Intel. And the goal of it was to, as you say, ⁓ provide a vehicle for which the community can receive high quality ⁓ quantum devices from established... ⁓ fabrication processes throughout the country. And so our role within it is to provide superconducting qubit devices in service of folks' to them so that they don't have to worry about the fabrication, the packaging, all of that. And so this has been something, as I said, it's been going for about four years. There's upwards of 50 users now. ⁓ And the way that typically looks is you submit a project application and as long as your project fits within some criteria that the sponsors want us to provide these devices to you, ⁓ you get a design rule checker ⁓ software package to make sure that the designs that you're generating for your experiment are going to, basically are they gonna yield? Are they gonna be able to be fabricated in our process? And you then submit those GDS designs to us. We aggregate them into a multi-user mask and our team fabricates them and then ships them out to the community. So, well, to the person who submitted them, I should clarify. But yeah, so, you know, it's now it's about 50 groups. Even early on folks were able to give presentations and write papers based on these devices and that number has grown. I don't have that number off the top of my head, but it's in the tens of papers that have been published based on devices fabricated through the Squirrel Foundry. ⁓ And that's also not including ⁓ any of the internal work from Lincoln Lab or the work from MIT campus ⁓ that has been using the same fabrication processes for many, many years in service of those experiments. yeah, so people use it for a variety of different ⁓ application models. ⁓ There are some groups who, you know, if you're a new professor starting out, and it's going to take you six months or a year or more to sort of get new fabrication equipment into your lab and installed and develop processes and be able to make, you know, reliable superconducting quantum devices. Submitting to the Squill Foundry is a way to really accelerate that onboarding process of starting up a new lab, for example. We also offer some advanced capabilities. So there are other folks who maybe even have a well-established process for making quality superconducting qubits, but maybe they never took the time to develop airbrush crossovers or flip chip capabilities. And so there's like another layer of complexity that we offer that can also fill a position within within an established research group's operating model. Additionally, tools go down, right? So having a parallel pathway to have devices fabricated is generally useful for any group. If... ⁓ you know, your eBimuthography system is getting a little old or something, you know. ⁓ If you're not confident in its reliability, ⁓ diversifying, ⁓ you know, the ways in which you are acquiring devices and fabricating devices can be really helpful for risk mitigation in those kind of ⁓ schemes. So the SQL Foundry and generally speaking, it's called the Qubits for Computing Foundries. ⁓ ⁓ with LPS, they do a lot of different things for lot of different people. ⁓ But there's really no doubt that it's making a big impact in the community and it's something that our team is again, really thankful to be able to be a part of and be able to support folks that way.
Daniel Rodan Legrain: I mean it is truly a fantastic program and 50 users that grew fast.
Kyle Serniak: Yeah, yeah, I mean there was a there was a initial like seedling year. It would be a year and a half when it started with maybe about 10 users and ever since then it's grown relatively steadily. We onboard new users approximately with every fab run that we do of which there are seven, eight a year that are for the Foundry users. And yeah, basically every, you know, we're constantly onboarding new folks with new.
Daniel Rodan Legrain: More broadly, how do you see Lincoln Labs' ⁓ in the quantum ecosystem compared to universities and ⁓ startups? Especially, I'm thinking in terms of enabling infrastructure, testing and evaluation, and supporting the broader community.
Kyle Serniak: Yeah, I mean, that's kind of singular to the role of any FFRDC in its respective community. so Lincoln Laboratory, I should say, is a much larger institution than just superconducting qubit research. But yeah, central to that mission is... Supporting the community and supporting the sponsor community in in all the ways that that makes sense. That's something that that varies field to ⁓ Typically, ⁓ you know, we're focused on advanced prototyping which which is an interpretable, you know ⁓ You know thing in and of itself ⁓ But you know, it it looks like academic style research done in a professional setting it looks like ⁓ a test and evaluation where our team is working alongside the sponsors to provide feedback on research. Another role that we have within the community is what's called government furnished capability. And for certain programs, the sponsors can, ⁓ you know, fund Lincoln Lab to support just very directly the performers on a specific program. And that can look like a bunch of different things as well. That can look like academic research, that can look like providing devices specifically for the goals of that program, ⁓ providing any expertise. It's an interesting role ⁓ that really relies on the unique relationship that FFRDCs have with the sponsored community. and is generally just a very nice and impactful way to engage with the community that isn't quite natural to do at an academic group or an industry.
Daniel Rodan Legrain: How do you balance standardization with the need for innovation here?
Kyle Serniak: That's a great question. you can't, you know, there's no point in standardizing without innovation, right? ⁓ If you standardize and you're really good at one thing and the next good thing comes along, you have to be ready to incorporate that. Otherwise, the work that you're doing isn't quite as relevant anymore. So we have really healthy efforts on both sides of that line. We really want to make sure that we, at any given time, can make really high quality qubits ⁓ for experiments that maybe, you know, way to describe them is anything that isn't fabrication development focused. Like we want to have something that we know is just going to work when we're designing a new qubit circuit or trying to develop a new metrology protocol or... doing something that's focused on the rest of the infrastructure, the measurement chain, the way we're controlling the qubits. So we wanna make sure that there's always a complete process flow, fabrication process flow that is extremely well vetted, extremely reliable, and that we know it's just gonna work. And then on top of that, once that foundation is there, We spend a lot of time doing materials R &D, processing R &D, trying to understand what is the next change that we can make to get a performance improvement on one axis or another. Maybe it's cubic coherence. ⁓ Maybe it's the reliability of the Josephson Junction critical current targeting. And within our group, This is something that's very central to our sort of identity is we have the full matrix in our minds of, is this going to affect this other axis? And do we need to worry about that? And we're constantly doing the vetting and validation to ensure that this one thing that maybe is great for one metric, is it going to come back to cause a problem? you know along some other axis. So yeah. No, no I'm rambling a bit I was just gonna say ⁓ that like to your original question you know you can't be stagnant you always have to be developing.
Daniel Rodan Legrain: Let me ask you one for... ⁓ sorry, Kyle. So let me ask you a looking forward question here, following up on what you said. And where do you see the biggest opportunities today to keep improving superconducting qubit performance?
Kyle Serniak: Yeah, I mean, I think it's you know, similar to what I was saying before There are all of these different avenues that one can pursue to improve qubit performance a You know a pretty you know, pretty exciting one is the materials Because basically as you improve the materials that we're using for superconducting qubits Basically everything else gets lifted by that ⁓ By that found it's it's kind of the foundation everything gets lifted by by those improvements in a similar way ⁓ The same can also be said for the you know novel qubit circuits where the information is encoded in different ways You can kind of just flip the logic and say ⁓ well for any given Material system if you can if you can encode information in a way that's less sensitive to the defects in those materials Well, everything is just going to get better even with the existing you know, construction methods and materials. The same can be true of if you just improve those, then everything just gets better. So the materials, would say, are really a big... There's a lot of opportunities for improvement and a lot of things that we know we're learning more and more what to avoid, how to avoid it.
Daniel Rodan Legrain: Let me switch gears a little bit, Kyle, before we go to the ⁓ question from the audience. And let's talk about the people that are building those systems. ⁓ And from your point of view, what capabilities or mindsets do you think are becoming essential for the next generation of ⁓ quantum hardware builders, let's say beyond traditional physics training here?
Kyle Serniak: Yeah. So, I mean, very much central to the notion that there are all of these different, you know, there are a lot of things that need to go right ⁓ when making supernetic quantum processors. And I think it's natural maturity of a field to where, you know, the knowledge that was maybe 15 years ago, condensed into everyone's head of a little bit of cryogenics, a little bit of micro-engineering, a little bit of fabrication engineering, a little bit of this, a little bit of that. It's very reasonable for it to be more distributed and that allows for much, much more depth of understanding of all those individual components. And so... For folks who are interested in getting into the field, there's still a lot of value in the general superconducting qubit educational sort of pathway, but there's a lot of space for people who aren't yet in the field. Folks who have done process engineering at... at industrial foundries for, you know, and built up experience that way. That gives, that gives a completely different perspective than someone coming with a PhD from super 90 qubits. ⁓ Similarly, know, professional micro engineers are amazing, right? They, they can do, they can, they, understand the problem and, and, know how to implement solutions just much more effectively than, than than I do ⁓ as a armchair micro-engineer maybe. ⁓ So whether it's cryogenics engineering, thermal management, micro-engineering, fabrication process development, ⁓ quantum algorithm, the field is becoming a little bit more specialized, but in a way that is good and natural for... ⁓ you know, for a field that's and immature.
Daniel Rodan Legrain: What advice would you give to people choosing between, let's say, academia, research labs and startups?
Kyle Serniak: I think, you know, learn as much as you can about the real differences ⁓ because in the field, ⁓ you know, everybody has a role to play and there's a lot of overlap, but, you know, thinking about it day to day, if you look at the situation that you're in, you know, the position that you're in and you see, ⁓ I really, really like this or I really, really don't like this. You know, it sounds a little bit silly, you know, do the due diligence to figure out where you can maintain the things that you really like and where you maybe don't have to do the things that you don't like. If you don't like teaching, maybe academia is not the right spot. If you really love research and nothing else, maybe all of those could be interesting options for you. ⁓ And it's more about the secondary things.
Daniel Rodan Legrain: So one last question from Mikhail. What do you wish you had been told as a first year PhD student?
Kyle Serniak: ⁓ well, that's a good question. So, so many things. I, I wish I had been told, this is actually something I've noticed a lot. So, so moving into a PhD, ⁓ and especially kind of the, middle years of a PhD, is the first time for some folks, like if you, if you haven't worked in between, like if you go straight through school.
Daniel Rodan Legrain: you
Kyle Serniak: ⁓ It can be the first time that you are in a situation where you're doing work and there's no longer sort of the six month or four month for a semester timeline associated with a project. ⁓ And that's something that people often get really bogged down with. The notion of open-ended research being something that can take years and years when You're used to going through school and taking a class and having a final exam or final project and everything being really well contained in this four month ish timeline. And as soon as as soon as you're out of that and really deep in the research, that's that's something that you you can structure in a way that looks similar, but it's pretty easy to to sort of, ⁓ you know, not have that structure. So getting used to getting used to the fact that research can be relatively long-term endeavor. ⁓ I'm not sure the right advice that I wish I had gotten in that respect, ⁓ but I think that's something that folks consistently struggle with, kind of in the middle of a PhD ⁓ that I've certainly noticed and felt myself.
Daniel Rodan Legrain: Thank you, Kyle, for this and all the other insights you shared with us today. ⁓ We have a few minutes for questions from the audience. I see quite a few here populating the chat. ⁓ I see quite a few questions about the squirrel.
Kyle Serniak: Of course. Yeah, no, thank you, Danny.
Daniel Rodan Legrain: program. Let me group them together into one. You know, if I'm a new PI, how can I collaborate with Lincoln? Do you take industry partners? How long does it take to fabricate and ship these devices? What do you have to say,
Kyle Serniak: Yeah. Sure. Yeah. Yeah, so if you are interested in working with us, so if you just Google Squill Foundry, ⁓ there should be a webpage that comes up with an email address that you can send. I believe it's squillfoundry.ll.mit.edu. And from that, you can get all the information that you need about becoming a part of that ecosystem. ⁓ We do have engagement models for all types of institutions, including industry. So don't let that scare you away, ⁓ wrote about that. yeah, in terms of collaboration, absolutely, it's something that we do. ⁓ It depends a bit on the circumstances and it needs to be something that makes sense for us to work more closely together on, but that's also something that we certainly do. ⁓ I don't remember, I think there was going be one more question in there, but...
Daniel Rodan Legrain: That was how long does it take to ship a device.
Kyle Serniak: ⁓ yeah, good question. So it depends a bit. It depends on, you know, whether tools are working as expected or not. Typically, ⁓ it's about eight weeks, eight to 10 weeks ⁓ for between submission of the designs and shipping of the devices. ⁓ Sometimes it's shorter, sometimes it's longer.
Daniel Rodan Legrain: I mean it's really amazing that you design a chip and then it comes to your mail without needing a whole cleanroom in your university.
Kyle Serniak: Yeah, it's, yeah, and that's, I'll say that is something that, you know, the timeline is longer than what a lot of academic groups can do themselves. And, you know, it's just a different way of working. It's not long, as I understand from a commercial foundry perspective. And... You know, it just, it changes the way you think about an experiment. You can't do as much serial development, ⁓ but you can really parallelize the things that you're doing and really answer questions that way. Different way of thinking, but similarly effective.
Daniel Rodan Legrain: Thank you, Kyle. ⁓ We have a student here who doesn't have quantum courses in his university and is asking, how can we learn about all you're sharing with us today without necessarily having to go to a classroom that I don't have access to?
Kyle Serniak: That's a great question. So the. I'm now forgetting. I mean, there's a lot of online resources specific to superconducting qubits or quantum information in general. I remember, you know, the IBM Qiskit materials being really, really helpful for folks onboarding into the field. ⁓ And there are a variety of other programs. Qubit by Qubit is one that comes to mind that's focused, I think, on a summer course on... ⁓ on, you know, that focuses on supergnetting qubits as well as more quantum information generally. But, you know, honestly, there's a network of folks. There's a network of folks, and if you reach out to them, then, you know, we have a pretty friendly community, and any connection that you have, will likely be happy to share with whoever they're... their favorite resources, the things that they find most valuable.
Daniel Rodan Legrain: I agree. Let me also mention that this Fireshot Chat series Quantum Builders is a good starting point to get to know who is available out there.
Kyle Serniak: Yeah, there you go. That's right. That's right.
Daniel Rodan Legrain: And Kyle, we have time for one more question. And you mentioned something that the quantum computer of the future might be different to the ones we have today. And I guess someone picked on that nuance and would be curious to hear yourself on how does the quantum computer of the future would look like.
Kyle Serniak: I don't, mean, so generally speaking, ⁓ I mean, it could, it could look like what we have today. I think people have, people have generally, you know, a lot of the optimism in the field comes from the fact that what we have is pretty good. And it's, the academic in me and the academic and a lot of people who, ⁓ who, who really think of themselves, yeah, there must just be, surely there's something a little bit better. And The fact that, especially for super-uniting quantum processors, the way you do control, the way you encode information, the materials that you're using, all of these things are really up to the experimenter, the designer, to really optimize and really understand and balance different trade-offs. I mean it more as a general statement that I'd be surprised if it was so similar to what we have today. ⁓ I think it's all of those avenues that there's a lot of room for improvement.
Daniel Rodan Legrain: Time will tell.
Kyle Serniak: Time will tell.
Daniel Rodan Legrain: Kyle, thank you so, so much for joining us today. I mean, your work really highlights something. Actually, we also really strongly believe at QBlocks that progress in quantum computing comes from understanding the underlying physics, building strong infrastructure, and enabling collaboration across the ecosystem. ⁓ As a small token of appreciation, we've sent you one of our QBlocks mugs with our motto, building quantum together. Hopefully a good companion for your next round of experiments. ⁓ For everyone watching, the recording will be available shortly on the QBlocks YouTube channel, on Spotify and Apple Podcasts under Quantum Builders. And please stay tuned for future Farsight chats. We have many more exciting gets ahead. And Kyle, thank you again. Thank you for joining us. Thank you for the amazing...
Kyle Serniak: Mm-hmm.
Daniel Rodan Legrain: insights and thank you for everyone joining us today.
Kyle Serniak: Yeah, thank you, Danny. This was a lot of fun. It was really great chatting with you and thanks to everybody for tuning in.
