Podcast | Manipulating Quantum Matter Online - with Infleqtion

Konstantinos Karagiannis: Have you ever taken a photo of interference? Grabbed a snapshot of quantum tunneling? Find out how to access quantum physics as a service in this episode of The Post-Quantum World. I’m your host, Konstantinos Karagiannis. I lead Quantum Computing Services at Protiviti, where we’re helping companies prepare for the benefits and threats of this exploding field. I hope you’ll join each episode as we explore the technology and business impacts of this post-quantum era.

Our guest today is the founder and chief strategy officer of Infleqtion, Dana Anderson. Welcome to the show.

Dana Anderson: It’s a pleasure to be here, Konstantinos. Thank you very much for having me.

Konstantinos Karagiannis: Thanks for coming. This special episode covers how listeners can get hands-on with the quantum world in a way they’re not used to. But before diving into that, tell us quickly how you found your way to quantum.

Dana Anderson: It was hardly an accident. I have physics in my genes somewhere. My father was a physicist. He worked with Enrico Fermi. In some sense, I was born into a quantum world. But my personal one started when I went to graduate school and I got a Ph.D. in quantum optics. And quantum optics has an awful lot of parallels to what we’ll be talking about today, with atoms instead of photons. My story starts, probably, in 1975, but goes all the way up to the present day. Fill in anything you like, but you might be more interested to figure out how we got into cold atoms.

Konstantinos Karagiannis: That’s great. And we’ve had Infleqtion on before. Can you give listeners an overview of the types of work your company does, in case this is the first episode they’re hearing?

Dana Anderson: Infleqtion was formerly ColdQuanta. It was founded all the way back in prehistoric times, 2007, following the first demonstration of— Bose-Einstein condensation. We wanted to get the technology out to the world, so we started making products that enable other people to get this fifth state of matter, as it’s often called.

By now, this company is hardly a startup. We do everything based on atoms, whether they’re atoms or ions — from clocks to quantum computers and many things in between, including quantum memory, including RF detection by atoms, which we call quantum RF detection, and inertial sensing is one of the things I’m very close to because I’ve been doing that since my Ph.D. — a very broad spectrum of quantum things and quantum components.

Konstantinos Karagiannis: And some terrific software capabilities too. I love the folks on that team as well.

Dana Anderson: Super.tech. They’re outstanding.

Konstantinos Karagiannis: Yes — great gang. Now, on to Oqtant. When we get hands-on here, it’s usually software-related. This is a little different. The closest we’ve come to something like this is that we had one episode where you were building your own quantum device, but it’s not a quantum computer. It was more like a neutral-atom space. This is the first time we’re talking about accessing something that’s not a computer on the cloud. Can you give us a sense of what the platform does?

Dana Anderson: The hardware platform produces a quantum state of matter called a BEC — a Bose-Einstein condensate. And the interest is what you do with a BEC. But even before what you do is understanding, getting a very different picture of quantum than you might have — hands-on. You can take a picture of a quantum wave function. But in the end, of course, you do talk to the machine with software — and software is going to play a very dramatic role in what the machine becomes, as we’ll talk about. But let me turn it back to you with questions, and we’ll go deeper and deeper as you wish.

Konstantinos Karagiannis: Keying on something you said we’re going to drill into in a moment: You’re literally able to take a picture of something happening in the quantum realm. That’s amazing. You initiate it, and then you capture this. I was able to see some of that visualisation back at the Q2B show, and it was pretty impressive. That’s why I wanted to have you on. Can you start with a refresher, or even an intro to some listeners, on what a Bose-Einstein condensate is?

Dana Anderson: Since we’re on only audio and not video, you can’t see my hands. I’m going to do my best to describe what my hands do. But my colleagues at the University of Colorado, Carl Wyman and Eric Cornell, and Wolfgang Ketterley from MIT, demonstrated the first Bose-Einstein condensation.

Not surprisingly, you know, Einstein’s name is associated. And an Indian colleague, Satyendra Bose, said when you get atoms close enough together and cold enough, they will undergo a phase change, much, in many respects, the same way as ice turns into water or water back into ice as you cool it — a phase change to a different kind of matter: purely quantum. When you get the atoms close enough together and lower the temperature enough, very close to absolute zero, they form a single quantum wave function.

Many people know that an individual atom will have quantum properties. But when you take a whole cloud of them — get them close enough together — they lose their identity and get together to form one quantum wave function. It’s called macroscopic because you might have many atoms participating, and they can be the size of almost the width of your hair. It is amazing. You can take a picture of them, and that’s taking a snapshot of a quantum wave function.

Konstantinos Karagiannis: That’s impressive because most people don’t think of anything remotely macroscopic as still being quantum. And, of course, we’ve had different things over the years that cross that barrier, like buckyballs, for example. You’re able to do quantum work with that. In theory, if we separate it from the device for a moment, why should someone care about a BEC? In theory, what can they do?

Dana Anderson: Let me move all the way forward and then back up. One would love to take what we do today with electrons — electrons are at room temperature and don’t remember themselves as being quantum for very long; it’s just a couple of picoseconds — and do many of those things in the quantum realm instead. You can think of the circuits you have in your iPhone, although they’re very powerful and very sophisticated, be a little bit more primitive— going back to the early days of transistor radio, say — and do everything that we do today with electrons, with atoms instead, atoms that are in the quantum realm.

You can make, for example, incredibly sensitive detectors. One of our interests is replacing positioning systems that usually use GPS with a sensor to tell you where you are and what time it is. When you do that in the quantum domain, it can be far more accurate than any other way of doing it.

In principle, you can make circuitry powered by Bose-Einstein condensates, roughly speaking, in much the way that you make electronic circuits these days powered by batteries. And that’s a little far-fetched. But that, in fact, is what our goal is and why we’ve started to put these machines on the cloud, so other people can become innovative along these lines. 
Now we can back up. I’ll pause for you, but we can back up to more immediate things — a little more palatable and tangible.

Konstantinos Karagiannis: There are some forward-thinking, theoretical hopes to come out of this. In a little while, I want to ask you about what you might have already seen people doing. What can you do now, let’s say, if you were to log in. What would you be able to experience and experiment with today?

Dana Anderson: Let’s first talk about from an educational perspective, because quantum is hard. Wrapping your head around what quantum is, is quite a challenge for most of us, including me. First, all those things you’ve heard about, such as superposition, such as tunneling, you can do all these experiments by starting with a cloud of atoms you cool down to about 100 billionths of a degree of absolute zero. You could never do that before — reach a machine and say, “I want atoms that are 100 billionths of a degree below absolute zero, form a quantum state and watch them tunnel.”

It’s still mind-boggling to me today that you can take a pair of clouds of these atoms, each of BEC, and watch them come toward each other and interfere. And we usually think of atoms like Ping Pong balls: if you had two clouds run into each other, they would wreak havoc and bounce all over each other. Instead, suddenly, when these clouds go, you see regions where there should be balls, should be atoms, and there are not — it’s dark — and others where there are far more atoms than you think there should be.

This is the phenomenon of interference. And the fact that you get Ping Pong balls, so to speak — atoms — to interfere is mind-boggling. The temperatures are mind-boggling. The interference is mind-boggling to watch. Then these strange notions, such as tunneling, you can see for the first time. That’s first getting your head around this quantum phenomena you don’t have access to normally.

Konstantinos Karagiannis: To pause for a second with the cloud idea, sometimes, when people hear cloud, they’re thinking of a bunch of places that a particle might be if you make an observation — the idea that instead of there being orbits around an atom, for example, you have more like a smearing. And then, somewhere in that cloudy ring would be where the actual electron is if you were to make an observation.

Dana Anderson: Remember, this might be 10,000 atoms. They form a cloud. We don’t know, and it makes no sense — this is the annoying thing about quantum mechanics — to ask, “Where are the atoms?” If you look, you’ll see them. That’s why you image them. But as long as you don’t look, they’re only acting as one entity we call the BEC, and not a collection of individual atoms. When I say cloud, I mean this collection of atoms. And I say cloud partly because they’re held in a vacuum, and they’re held in place by magnetic fields and laser beams, and you shine light through them. They look like they’re just sitting there. In that sense, they are cloud, and they take the shape of the container, but the container is made of light.

And let me tell you, that’s the beautiful thing. When you are programming this, you are programming how you’re shooting laser beams at the atoms, and you’re making them behave with laser light. And we’ll get more into that, perhaps. But that’s another remarkable thing —they’re just sitting there in the middle of a vacuum system, but all controlled with laser beams and sometimes magnetic fields.

Konstantinos Karagiannis: Then you see this, and you can capture images of it, and then you can do the interactions, like you said. We could see them interfere, and then we could see tunneling happen right before our eyes.

Can you describe, since we can’t show images, what tunneling would look like if someone was using this?

Dana Anderson: You can, so to speak, put atoms in a bowl. You keep them in a bowl, ] if you think of a bowl of water, that water takes a certain level, depending on how far you fill the bowl. And there’s a lip of the bowl, and you can literally make a lip of a bowl that’s done with a laser beam instead of a massive material. It’s made with a laser beam — you make a wall and a lip of a bowl. And even though the lip is above the surface of the liquid — the surface of the bowl of atoms — they show up on the other side. That’s called tunneling because they shouldn’t be able to get over it, but they can. The reason they can, and do, is because they behave as a wave, not as particles you need to throw over a wall. They can partly go through the wall, like light beams go through partially silvered mirrors that light beams will partially get through. The atoms will partially get through the wall. This you can take snapshots of and say, “They shouldn’t be on this other side, but they are.”

Konstantinos Karagiannis: Imagine, instead of spending time trying to create diagrams to show what quantum effects might look like, if we can imagine them, we can actually see them happening. That’s what I thought.

Dana Anderson: You take snapshots like birders like to do with rare birds. You can say, “I took a snapshot. And here it is.”

Konstantinos Karagiannis: The ultimate PowerPoint. That’s amazing. I wanted to clarify that for everyone listening.

These are some of the types of basic experiments we can expect. Let’s say we’re doing something like quantum tunneling. Do we get to do precise measurements and see how many atoms went across?

Dana Anderson: You can. The instrument, by some standards, it’s the same in quantum computing. There’s no great quantum computer out there. We’re expecting later. The data, you can certainly take. You can certainly count the number of atoms. But it will be noisy compared to some perfect picture one has. It’s partly noisy, in fact, because of quantum disks — the atoms are quantum themselves. When you take a picture, you’re actually looking at atoms, and maybe it’s not surprising that’s a little bit noisy. But, yes, you can take data, and as time goes on, this machine will have more capability, and you’ll be able to take better data, but you can take some today.

Konstantinos Karagiannis: What’s amazing about that is, even though there’s noise, these are the building blocks of translating from the theoretical ideas of what happens in that subatomic world to actually being able to count and see and maybe even one day compare an expectation using something like a wave function of what percentage you expect to cross and what actually crosses.

Dana Anderson: You’ll get all those answers about right if you do the calculations. In fact, what you didn’t see at Q2B is our upcoming simulator. You will be able to calculate, using conventional classical computers, what you should see. In simple experiments, you’ll see the same thing as the calculation, or pretty — at least resembling pretty — well.

As time goes on, the fact is, the machine can do things you just can’t calculate. That’s the nature of quantum. The idea of the simulator is, you can get an idea, but then you’ll have to turn to the machine for what really happens, including when you want to do entanglements and things like that. That’s the fascinating thing. But there will be a simulator so you can do a sanity check. The simulator should tell you what you expect, and then you’ll see that the atoms do pretty much that in simple experiments, but a noisy version. As time goes on, they’ll be more faithful. And, as time goes on, you’ll find you can do experiments the calculations just can’t predict.

Konstantinos Karagiannis: These experiments, are they fully documented in the interface so someone who’s listening can go in and do what we just described, for example?

Dana Anderson: Oqtant was first publicly announced in December. Unlike quantum computing, where there’s been this very long history, since the 1980s, of people thinking about it, much less time has been put in there. Yes, there are document experiments about doing interference and something called the Talbot effect, where you can see the atoms interfering. Over time, there’ll be more and more, such as tunneling experiments. They’re coming, but it’s a young time now, so we invite people, in fact, to write programs, leave them with us. There’ll be open source coming in time — an open source way to do that — so people can share what experiments they’ve done and the programs they’ve done them with.

Konstantinos Karagiannis: Just like models we’ve seen in any kind of programming platforms for working with, let’s say, a quantum computer, where people share their code. But in this case, remember, you’re actually seeing these physical things happening, which is mind-blowing — all these years I’ve spent imagining it, from blackboard to whatever, now to see it’s real.

Dana Anderson: And they do all the things you’ve learned, and then some.

Konstantinos Karagiannis: Can you explain what users would get with, let’s say, the free versus paid tiers?

Dana Anderson: You get 10 shots per day for free. There, you have full access to the machine. You can program either via the web interface, which is pretty primitive in what you can do, or by Python. Octopi, we call the language, and you can program what it does in Python. You get the data out, and you get 10 shots.

The machine is on for several days a week and then off to upgrade or check and so on. But you can submit a job anytime. But it runs, obviously, only when it’s up, so you don’t know, necessarily, when it’s going to come out after you put it in when you’re working for free. When you have a nominal subscription — about $10 per shot — you have priority in the running. In the free version, if there are other people that have put in a lot of jobs before you, you are where you are. The paid jobs are prioritised.

If you’re a hardcore researcher, you probably want to call us and get a subscription because you might want thousands of shots and you might want to basically own the machine for this slot of 10 minutes so you get absolutely reproducible results, things like that. That’s not articulated — except to say give us a call, because we want to meet people’s needs. It’s the first for anybody who walks up as either free or use a credit card. After that, if you want to get pro, it can get quite sophisticated about what you do and how you’re given time to do it.

Konstantinos Karagiannis: It’s still a great value because I assume this thing cost a few million to make.

Dana Anderson: And to run and maintain. It’s a couple-of-million-dollars machine. But more importantly, it’s the time and the number of Ph.D.’s it takes to build one of these machines and maintain it. That’s what you’re paying for. Ten dollars may seem like a lot compared to going to Amazon Web Services, but take my word for it — for now, it’s quite reasonable. Our hope is that masses will eventually use the machine and the price will come dramatically down.

Konstantinos Karagiannis: When IBM first put quantum computers on the cloud, we saw the papers start pouring into arXiv. Has this platform led to any papers yet?

Dana Anderson: It has in a private use. But remember that when quantum computing went into the cloud, the world had already been invested. IBM, Google, had already invested considerable time in developing infrastructure and so on. First, users were primed. Here, our user community, this is brand-new.

There is a community out there that has used machines like this. Today, operating since 2018, is a predecessor to Oqtant, operating on the International Space Station for this many years. Since 2018, every day, it generates BECs, and papers have come out of that. Other researchers have used Oqtant-like machines for studying black holes, for example — very scientific research. Our hope is to enable an industry like the maker community of people inventing things. That’s what this machine is for — not yet. It’s been publicly announced in December. But we hope this is what happens, and we will encourage people to reach out to us. We’ll help them do that.

Konstantinos Karagiannis: This is the ground floor here as far as having access to something that can lead to novel research. What would you like to see this platform lead to or evolve into?

Dana Anderson: In my own domain, we use a platform like this. Later, I expect this will also be in the cloud to make sensors for position navigation and hardcore, serious GPS placements. The next thing is heavily intermixed with machine learning. I don’t want to get carried away with all the AI and so on, but quantum is hard to understand and hard to get to behave the way you want. I’m very much involved with machine learning approaches to controlling quantum systems to do useful things.

Sensors are one thing. The other thing I mentioned earlier is much more along the lines of what we call atomtronics —the atomal analog of electronics. One can make atom transistors — transistors that use atoms instead of electrons and work within the quantum domain purely rather than in the classical domain. You can make amplifiers, oscillators — you can combine the behavior, and this does get carried away, but just by analog, all these things go into a radio. I’m not suggesting you make a radio out of atoms, but there are all kinds of detectors and radio-like receivers, for example, one can make using atoms instead of electrons that are incredibly sensitive.

You can also do the things like you do with compute — do logic, although if you want to do logic, a quantum computer is better. But there are other signal-processing tests you can do with Oqtant that you can’t do with the quantum computer that I look for people to do.

Konstantinos Karagiannis: With all that signal transmission, maybe even tiny electronics can benefit. Any nano devices might be able to have receivers that are smaller — take up less space, less power.

Dana Anderson: Like the compute, there’s likely to be functionality that is hybrid. You have a quantum system over here. It’s processed quantum. For example, you’re not going to do communication with atoms. They’ll always translate into light. You’re not going to do heavy-intensity classical computing with atoms. You’re going to leave that to electronics.

I envision atomtronic circuitry side by side with electronic circuitry to carry out very sensitive detection, signal processing, and then make it useful to the world by communication or classical processing.

Konstantinos Karagiannis: Can you envision any other experimental tools being made available as services? Is this the first of many types of things you might want to consider offering?

Dana Anderson: There’ll be more and more functionality given more and more components. But every component with atoms is made with light. That’s something I want to emphasise here: If you build an atom circuit and you don’t like it, you just rearrange the light by your Python programming. It’s the functionality. You’ll be able to put more and more devices, more and more complex light patterns, to make more and more complicated patterns — and, as the community demands it, more and more functionality. For example, if you want to do RF signal processing, maybe there’ll be an RF receiver that feeds into the atoms added to the machine.

Konstantinos Karagiannis: I was also thinking, could this be the beginning of a new breed? You’re working on the gravimeter that detects anomalies. Would there be versions of that that might come online so people can experiment?

Dana Anderson: Absolutely. That is the other thing I was describing that I hope will come online — the development of inertial sensors for measuring Earth’s gravity. The scenario would be that an engineer or an inventor creates a new kind of sensor — maybe more precise, maybe it senses subtle seismic motion of Earth, or strange r fields. Let them design it there, then it gets turned into something that’s actually deployed and used. It is a different culture of invention.

Infleqtion’s intention with Oqtant, aside from the educational one, is to get many more creators and inventors access to very difficult technology and make it useful to speed the innovate, design, deploy cycle that quantum so much needs and is such a barrier to because it’s so hard to make. This is designed to be a new paradigm for creation and innovation in the quantum domain.

Konstantinos Karagiannis: It’s amazing. The whole maker approach, when I think back to the early days of “x as a service,” it was always infrastructure, software, that’s it — whatever. Now, we’ve reached this level where we’re playing with atoms as a service, one way or another.

Dana Anderson: It’s quantum mechanics as a service and atomtronics as a service. Then, to turn that into reality in time —making, delivering things based on the designs of users of Oqtant. A key point is, users of Oqtant, anything they do as theirs. All the IP is theirs. That’s meant to generate interest in innovating with the platform where people get to keep their IP, develop their technology, and be able to leverage online access to quantum mechanics and quantum systems.

Konstantinos Karagiannis: If you’re patient, there’s nothing stopping you from going to the link in the show notes and giving this a try — and everyone should, just to see it. It’s amazing. The first time I saw it, I was blown away. I just knew the gravity of what I was observing.

Before I let you go, are there any other updates on Infleqtion you want to give? Your roadmap, or other kinds of things?

Dana Anderson: A roadmap for Infleqtion is going to be coming through a webinar in the not-too-distant future. We’re talking about this in the following week. You’ll hear much more about our computing effort. We’ve been pretty much below — not very much above — the radar in what we’re doing in quantum computing. And a key aspect is that what you see in quantum signal processing, is very much related to what we’re doing in quantum computing in the long term — and, in particular, to get quantum technology at the edge, as our world likes to refer to it, you have the sensing, the signal processing and the compute all very much at the edge in the future.

Folks will get a picture of that. You’ll make me happy if your audience uses Oqtant because frankly, we have to show that it’s useful. We want to see people doing useful things with it and bring quantum to the masses. And keep your eye on it, because it will be upgraded much more quickly than you see quantum computers upgraded and have more and more capability over time. Please do come back.

Konstantinos Karagiannis: You heard the call to action right now: arXiv is a plain field waiting for your papers.

Dana Anderson: And we will have competitions to make the coldest atoms, the most atoms and the most interesting circuits with atoms. Watch for those as well.

Konstantinos Karagiannis: That’s great. Dana, thanks, again, for coming on.

Dana Anderson: Thank you for having me. All the best to you, and all the best to the community we’re talking to.

Konstantinos Karagiannis: Now, it’s time for Coherence, the quantum executive summary, where I take a moment to highlight some of the business impacts we discussed today in case things got too nerdy at times. Let’s recap.

Infleqtion started 17 years ago, following the first demonstration of Bose-Einstein condensates. What are these BECs? Basically, when particles get cold enough and close enough together, they act as one wave function. BECs can be quite large, relatively— almost the width of a human hair.

Since 2007, Infleqtion has worked on many practical implementations of quantum technologies, including quantum computing and sensing, atomic clocks, and software implementations we’ve covered on the show before. But now, they’re making available online a special device, Oqtant, that lets you interact with BECs and see quantum phenomena occurring. As fascinating as this is to look at in the web interface they provide, picture little colored dots swarming and interacting clouds. You can perform real experiments with predictable outcomes. You can verify everything from interference to quantum tunneling.

Infleqtion hopes Oqtant will create a new quantum maker industry. Will some of these experiments lead to better sensors? Will we see circuits made of atoms that can help miniaturise other technologies? It’s an open field of experimentation, and you can run 10 shots a day on Oqtant for free. Check out the link in the show notes.

That does it for this episode. Thanks to Dana Anderson for joining to discuss Infleqtion, and thank you for listening. If you enjoyed the show, please subscribe to Protiviti’s The Post-Quantum World, and leave a review to help others find us. Be sure to follow me on all socials @KonstantHacker. You’ll find links there to what we’re doing in Quantum Computing Services at Protiviti. You can also DM me questions or suggestions for what you’d like to hear on the show. I’ll be gathering these for an AMA episode I’d like to do in the future. For more information on our quantum services, check out Protiviti.com, or follow Protiviti Tech on Twitter and LinkedIn. Until next time, be kind, and stay quantum-curious.