Interview by Audrey Russo
Quantum computing has spent years being described as the technology that’s always 10 years away. In Pittsburgh, that future is getting a lot closer.
University of Pittsburgh Senior Vice Chancellor for Research Rob Rutenbar joins Pittsburgh Technology Council President and CEO Audrey Russo to unpack the quantum computer headed to the Pittsburgh Supercomputing Center through funding from the National Science Foundation.
And don’t worry if you can’t tell a qubit from a Q-tip.
Rutenbar starts with the basics, explaining how quantum computers work, why qubits can exist as both zeros and ones, what Einstein meant by “spooky action at a distance,” and why Pittsburgh’s new Rigetti system has to operate at temperatures colder than outer space.
The PSC’s nine-qubit system is considered a starter quantum computer, but its arrival gives Pittsburgh researchers something enormously important: hands-on access to an emerging computing technology that could ultimately tackle problems conventional computers struggle to solve.
Hit play to discover why Pittsburgh is building serious momentum in quantum computing, how quantum could transform cryptography, drug discovery, materials science and optimization, and why pairing a quantum computer with PSC’s world-class supercomputing infrastructure creates one very powerful combination.
Quantum isn’t just coming.
It’s coming to Pittsburgh.
Transcript:
quantum
~We're here in the Huntington Bank Studio here with the Pittsburgh Tech Council. They've been long sponsors of ours and we've been doing podcasts and having a chance to talk to really thousands of people.~
~And this is pretty special though, because having a chance to talk with you, do a deep dive, you've served on our board, but I think more importantly than that, you've had a really interesting, journey in terms of being in Pittsburgh, leaving Pittsburgh, starting companies, coming back - Coming back to Pittsburgh.~
~Coming back to Pittsburgh. Yeah. Being at the helm of, a lot of the research and innovation over at the University of Pitt dabbling in yourself and companies. So I can't think of a better person to have a conversation with about quantum. Yeah. Happy to be here. Always happy to be able to talk about wonderful, nerdy stuff like quantum.~
~I think it's fantastic. So let's, first of all, let. I want everyone to know just a little bit about like the role that you have now even though that role doesn't define - Sure. All of you. I am senior vice chancellor for research at the University of Pittsburgh. The typical term of art for my job is VPR, vice president of research, right?~
~I am. I report to the chancellor, Joan Gable. I was hired by Pat Gallagher, right? Previous - I remember. Previous chancellor who, because he has a physics degree from Pitt knows even more about Quantum than I do. Oh, no. Should I have invited him instead? Not necessarily, but I'm just saying, the joke is that I now know more about quantum than anybody on the leadership team at Pitt, and that was not previously the case.~
~Oh. So, I am, I'm basically responsible for the care and feeding of the research enterprise. So we're a kind of a $1.5 billion research platform, so- So small. Yeah. Yeah. We, so I mean, I own the awards ecosystem for like writing proposals and stuff. I own the compliance ecosystem. If your research involves sticking a needle in somebody's grandma, you need to talk to my team to make sure that we're doing that okay.~
~I own the innovation ecosystem, like patents, IP, startups, investing and that sort of thing. I own the research computing enterprise. So, and we'll talk about this, the Pittsburgh Supercomputer Center - Right. On the Pitt side reports to me. And, we both make sure that the that trains are running on time.~
~And when you have umpteen,000 faculty and umpteen thousand students and umpteen,000 active research awards from umpteen thousand sponsors, the keeping the trains running on time is a pretty big deal. But, we also have a strategic focus. And one of the strategic focus areas that I've been leaning in on is making sure we have some game in quantum,~[00:00:00]
Welcome back to the Pittsburgh Tech Council podcast. I'm Audrey Russo, President of the Pittsburgh Technology Council, and today we're diving into one of the most exciting developments to hit Pittsburgh's tech and research landscape in decades. If you've been following the conversation around quantum computing, you know it's been called the technology that's always ten years away, until now. Well, Pittsburgh is about to prove that it's not ten years away anymore. It's happening right here, right now.
Our guest is Rob Rutenbar, Senior Vice Chancellor for Research at the University of Pittsburgh. Rob oversees a one point five billion dollar research enterprise, everything from compliance to patents to the Pittsburgh Supercomputing Center, which is one of only four leadership class supercomputing sites in the entire country. And he's been on a mission to make sure Pittsburgh has real game in quantum.
The National Science Foundation has awarded a [00:01:00] grant to bring a quantum computer to the Pittsburgh Supercomputing Center. The machine is built by Rigetti, a quantum computing company that traces its roots to IBM and Y Combinator. And it uses technology that seems straight out of science fiction. Qubits that can be zero and one at the same time. Entangled particles that Einstein himself called spooky action at a distance. And a device called a chandelier that operates at temperatures one hundred times colder than outer space.
In this excerpt, Rob breaks all of this down in a way that's accessible, engaging, and honestly, a lot of fun. Whether you're a quantum physicist or you're hearing the word qubit for the first time, you're going to want to hear this. So let's get into it.
Right?
Which is what we're here to talk about. No, I think, yeah. It's fantastic and it's great to have you here. And then, with all the changes at the federal level - there's just been such huge impact and it still seems to me [00:02:00] from where I'm sitting, right, from afar, and Joan's been a great friend to us as well, is that you're still doing an incredible amount of work despite the fact that things have shrunk.
Yeah, I mean, there have been some challenges, but the good news is that there are some very strong federal strategic sort of, priority areas. And AI is one. Quantum is actually, one of, one of, one of the others, and we have very s - we have risingly strong game in quantum.
So, okay. So, we're gonna have a variety level of people who understand - Yeah. Quantum, right? Yeah. So we're gonna. So let's, like, get to the kindergarten level first. Okay. Okay. So we got a grant from the National Science Foundation to buy a quantum computer, right? So first, where did it go? It's going to the Pittsburgh Supercomputer Center.
So that's a, a national supercomputer center with an incredible track record. It was started 40 years ago by faculty from, jointly [00:03:00] from Carnegie Mellon and Pitt. It was a big proposal from two remarkable faculty members, one at Carnegie Mellon, Michael Levine, one at Pitt Ralph Roski's, both physicists, computational physicists who wrote a proposal to buy a $45 million supercomputer, a, a cray for people old enough to remember what a cray is.
Wow. Okay? And the whole point of the National Science Foundation program was supercomputers are incredible assets. Nobody has enough resources. No individual university can just like, write a check to do this. So the feds will step in and essentially anoint a small set of centrally strong places to have some supercomputers, and then we'll put them.
We'll stitch them together into sort of a national network where people can people can get access to these machines. And so we got our first ever cray. It went in Westinghouse's [00:04:00] site, right, out in Monroeville, because that's where it had the sort of the, the infrastructure to support all the power and the cooling and all that other kinds of stuff.
And it's been just a, a spectacular asset for the region. There are only four of these things nationally that are at this level. So, one of them is at UC San Diego. One of them is in Texas, University of Texas, called TAC, the Texas Advanced Computing Center. One of these is at my former place of employment, University of Illinois Urbana-Champaign National Center for Supercomputing Applications, NCSA, and one of them is PSC, the Pittsburgh Supercomputer Center.
So there's only four of these things. There's the, like, at scale places with leadership class machines that, that people from around the country can get access to. So we wrote the PSC team wrote a proposal to the National Science Foundation for a quantum machine. All right, so what's going on there?
So the, the quantum is a little bit. Quantum computing is a little bit like nuclear fusion. It's this wonderful [00:05:00] technology that has always been about 10 years away, except now I can write a check and I can get one and I can decide how big of a one I want. And they're still very expensive and very experimental, but poised to dramatically reshape, the overall landscape of computing.
And so here's why. Regular computers, conventional computers operate on ones and zeros. Everybody probably knows that. And so the things they're built out of are bits, binary digits, right? Quantum computers are built out of things called qubits, kind of quantum bits.
And the thing that's weird about a qubit is that it can be both a one and a zero at the same time, and you don't know which, right? And it is just in the nature of the quantum stuff. And these things are. I mean, there are quantum computers that are made out of atoms as the quantum. There are quantum computers that are made out of interesting arrangements of semiconductors.
[00:06:00] Ours, the one we're getting is made out of a technology called Josephine Junctions, which are two superconducting thingies separated by an atomically thin layer of something, which then allows electrical current to pass seamlessly through it. And turns out that turns a micros - an atomic level quantum thingy into a macroscopic thing that you can see and control with circuits.
There are qubits made out of photons. There are cubits. I mean, it's like which, which are wonderfully called flying qubits. I can't make this - Oh wow. I can't make this - I can't make this up. I can't make this stuff up, right? I was in a meeting with one of our, one of our physicists once and he was talking about like, I wish we had a little more work in flying qubits.
And I went, he went, he looked at me like, like, like I just fell off the turnip truck and he was like, "Photons." Okay. Okay. I'm sorry. I'm slow. Right? They're made out of all, all sorts of interesting things. We have one made out of semiconducting semiconducting stuff. They are. There's two wild things going on here.
They can be zero and one simultaneously. [00:07:00] All right? That gives them an interesting new sort of amount of computing power, right? And, but the thing that's weird about it is like, well, how do you know it's a zero or one if you don't, right? And also they are probably zero or probably one, right?
And so, and it's not like a, if I flipped a coin, it's half heads and half tails. Like they can be induced so that - Oh. It's 90% of zero and 10% of one. So they can be morphed. They can be tickled, right, to do interesting things. All right. Well, how do you know which one it is? It's like, well, you don't.
It's 90% of zero and 10% of one. But if you measure it, if you sort of poke it and go, "Hey, qubit, what are you?" It will, and the word is collapse into a zero or one. And nine times out of 10 when you run this experiment, you'll get a zero and one time out of 10, you'll get a one. Okay, but wait, there's more, right?
You can entangle qubits. And entanglement is the next great weirdness of quantum mechanics. [00:08:00] It's the thing that Einstein literally called spooky action at a distance, right? Because it doesn't matter how far apart they are, right? When you entangle two qubits, they are not independently deciding if they are a zero or a one.
So it's not like the one on the left is a zero or one with some probability and the one on the right is a zero or a one with a different probability. It's that they're jointly two bits of two zeros, a zero and a one or a zero and a one or a one with a combined probability. And if you poke the first one, it changes the second one.
And it doesn't matter how far away it is either, right? That was the thing that Einstein said, like spooky action at distance. Turns out it's real, right? And so you can entangle them and b - and when you entangle them, you can compute multiple things in parallel, right? And so the computer that we've got from a company called Riggetti as we talk about the story of like howRigetti became Riggetti, right?
Yeah. It's nine qubits. It's a starter [00:09:00] quantum computer which is still - Like sourdough. Like sourdough, right? It's very you keep it in a dark place. We can actually - yeah. There's an analogy there. I shall come back to the sourdough analogy. Exactly, see? Like sourdough, they are incredibly sensitive to the environment in which they live.
See, my analogy stands. So it is nine qubits. It's a three by three array of entangled qubits, right? And the entangled qubits are these, these these semiconductor things called Josephson Junctions, right? The guy named after the guy who got the Nobel Prize for sort of discovering the physics that made them that kind of Josephson, English physicist.
They they live in a bucket of liquid helium, right, at, in a thing called a dilution refrigerator that itself costs like a million bucks, right? They live at a few hundredths of a degree above absolute zero, a few hundredths of a Kelvin, which is the physicsy temperature scale, around 270 degrees below zero, which is, which, which is, [00:10:00] I just got this off the website, 100 times colder than outer space, right?
And they need to be that cold because like sourdough qubits are incredibly sensitive to upsets in the environment, right? They're incredibly sensitive to temperature. They're incredibly sensitive to electromagnetic, like, things pinging on them. They're, you jiggle them a little bit. Literally, you look at them funny, right?
And they stop behaving properly. So they live in a bucket of liquid helium. The thing that is the quantum computer doesn't look anything like what you would imagine a computer looks like. It's literally called, it's called a chandelier. And so it looks like it should be hanging on the ceiling in Versailles, right?
It's the sort of, it's sort of a central cylinder with this sort of like set of concentric rings that get smaller and smaller. The things that are the coldest go on the bottom, the things that are sort of like near room temperature go on the top and you basically, you descend this thing into this sort of liquid helium cooled thing and the sort of the really crazy quantum stuff is on the bottom and the circuits that sort of plug into the rest [00:11:00] of the universe are on the top and it goes through layers of things.
We are waiting for the delivery of our quantum dilution fridge, which are because quantum is such a hot thing now are on back order. It's hard to get one. So like, so the supply chains for dilution fridges are a challenge, but because - That sounds like a good business to be in. It is a great business to be in.
They cost about a million bucks a pop, half a million to a million bucks a pop. But our partners at Righetti were very confident that we were gonna win this thing, and so they ordered one even before they actually knew we were gonna get the money. Oh, I love that they had inventory. I love planning. I love planning.
And so we're waiting for the delivery, we're waiting for the delivery on that. Because they are both zero and one at the same time and because you can entangle them, you can explore parallel solutions in, in, you can explore all of the solutions at the same time. So, so, the first qubit can be in two states with some probability.
The second one, two states, two times two, nine times 512. It can explore [00:12:00] 512 solutions at the same time. All right, that's good, right? The big quantum machine from Rogetti is 108 cubits. Okay? It is 12 ar - it is 12 repetitions of the three by three array. 12 times nine is 108. It's a four by three grid of three by three grids of Joseph's injunctions.
Two to the 108 is a number more than the number of atoms in the observable universe, right? That's how many solutions you can explore simultaneously. And that's sort of like where the power from this stuff, for this stuff comes from. True. And so we've got the starter computer, right, which is incredibly awesome.
And it turns out that setting up the problems for these things needs a regular old supercomputer next door. Oh wait. I've got one of those. Actually, I've got several of those, right, at at the PFC. So it's all completely freaking awesome. So, woo, I just went on a magic car- carpet [00:13:00] ride.
That's pretty magical. Yeah. So what does that mean for the world right now? - And the other things I wanna throw in there - Yeah. Is why have the last 10 years changed? All right. So, there are problems that are incredibly hard for conventional computers to explore, right? So these are the problems that are called like exponentially difficult problems.
They're the ones where you have to explore, billions and billions, quintillions of of, like, solutions to sort of to find the right answer. The thing that's wonderful about quantum computers is you can, like, it is the nature of qubits and entanglement that you can build algorithms on top of these things that explore those things all at the same time.
And the first thing that got everybody excited was cryptography, right? And so walking that back, most successful cryptography systems where you take the message, you encrypt it in a way that, like, codes it up so I can't read it, you transmit it across whatever the channel is, [00:14:00] and you decrypt it on the other side.
They're based on an encoding that is intrinsically some sort of a hard mathematical problem that you can't unravel, invert, undo, guess, right? And one of the most famous ways of doing that is a scheme called RSA named after Ravesh Shamir Nadelman, bunch of mathematicians at MIT. And it's based on the fact that factoring a number into prime factors is hard.
So prime numbers, right? Prime numbers are numbers that only, can be only divided by themselves and nothing else. Three is prime, five is prime, seven is prime, nine is not is three times three, 10 is not, two times five. There are cryptography schemes that are based on. So look, it's not hard to factor 10.
It's hard to factor a number that's 2,000 bits long, right? It's real hard to factor that number. And if I know the factors that make up that number, it turns out I can know some of the factors, you can know some of the factors, I can encode it, you can uncode it, but if you don't know [00:15:00] the stuff on both sides, you can't uncode it.
There are a lot of cryptography schemes that are based on intractably hard problems to solve. Quantum smashes those things like that. One of the first famous algorithms for a quantum computer was factoring things into primes. And it's just like Tuesday in the office, wham, right? So it breaks everybody's cryptography on day one when these things get big, which has caused a bunch of research in the community to look at quantum resistant cryp- crypto, which is like now a thing, right?
So, there are a whole bunch of other problems that in, in, in optimizing things in FinTech, right? In the simulation of physical systems. So there's a very famous quote from Richard Feinman, one of the, the great physicists of all time and a, an incredible character, that if we're really serious about being able to understand physical reality, at some point we're gonna have to be able to simulate quantum stuff because atoms are [00:16:00] based on quantum stuff - Right.
And the way electrons orbit is based on quantum stuff and chemistry is all basically the interaction of quantum stuff as atoms decide they do or don't wanna stick to each other. We simulate that stuff now by basically modeling the quantum mechanics and then dumbing it down because it's so crazy expensive to explore all of those paths, and we use supercomputers to do those kinds of things.
If you have quantum computers, you can just do that kind of stuff. So you can do drug design, you can do material design, you can do all sorts of amazing things, but you need more than nine qubits, right? But if you have a couple hundred or a couple thousand, you might actually be able to do that.
So there's a range of problems that are intractable on conventional computers. We find ways around to do them. Quantum just smashes those things and makes it possible to do it like that. That's huge. So what ~do you think that means for business and for innovation? - And what are we - Fintech will be very inter - Fintech is super interested in that.~
~You're already seeing AI come in and sort of apply layers of sort of the, the pattern management - Right. Pattern making, kinds of things to simplify sort of, rational drug design, molecular design. If you then have quantum stuff to like really simulate, I mean, imagine you have AI saying, "I think this molecule might do what you want," and then you have sort of a quantum computer that can like actually simulate how the molecule binds to the surface of a virus or something like that in a lot of fidelity.~
~Wow that, that changes the universe in some really big ways. And so what about digital twinning in terms of where this is in this space? Sure. I mean, there are a lot of optimization problems that that have, like many solutions, right? And if you can sort of formulate this in a mathy way of like, I just want, on this function of a million variables, I want an arrangement of those things that makes this particular mathy function a big number or a small number.~
~Yeah. I mean, there are, there, there are people in that community looking at how you take sort of more abstract versions of problems, take them over into the math and then take them sideways into into quantum optimizers, quantum solvers. Yeah, and I'm also thinking about personalized medicine.~
~Sure. I mean, anything that allows me to analyze the, at a molecular level why you were you right is certainly gonna be - And what you might respond to or not, right? Respond to. You could be a - Right. Because when we look at medicine, right? I mean, when we look at these - Yeah. Solutions, we are just not necessarily looking at it in a personalized way.~
~Right. Well, and I mean, you're probably tracking the, if it would be nice if you own some Merck stock or Moderna stock, over the last couple of days because they've just been announcing - Wait, did you give me an insight? Nah. It was, it's all over the news. I subscribe to the Financial Times and The Economist, so it's unavoidable.~
~They are getting positive results from their mRNA-based cancer vaccines. Right? Yeah. And so it may well be the case - Which mRNA also. Right. Right. Yeah. I was just gonna say, it may well be the case that we do not cure your cancer - so much as we vaccinate you against it so that it can't happen.~
~Right. That might be what happens 20 years from now. And that's completely wild. Right. Isn't it wild? Yeah. And so - But through replicating who you are - Sure. Being able to understand that. Yeah, you were able, we were able to build therapeutics based on RNA, right, that is highly highly tuned to you.~
~I mean, the thing that makes cancer difficult, a thing that makes cancer difficult is that cancers are highly personal. Right. Right? They're essentially diseases of your genome, right? So something in you replicated wrong right? And suddenly you got something - It's not a contagion that I got from you.~
~Yeah. Yeah. It's not like, polio was a virus. Right. And if we can whack the polio virus, you're good. Right. Right? Your pancreatic cancer might not be the same as my pancreatic cancer. Exactly. Might not be the same as his pancreatic cancer. And so that's where this stuff gets crazy interesting. And so have you seen any proliferation of companies interesting that are saying that they're in the quantum space and they feel like - There are.~
~Materially, they're - There are, there, there's a rising set. I mean, right now, we are in a. Let me answer your second question to the - Okay. Which was awhile ago to get to this one. What's different? For a long time, quantum computing was science fiction. Right. Right? Until it became engineering, right?~
~And I have a lot of engineering degrees, so I'm in favor of this - Thumbs up. Of this kind of - go engineers, right? When it stopped being a, a set of touchy-feely sciencey things, right, and started being a set of e- extremely difficult engineering problems, like, oh, we are rather good if you keep grinding on stuff and you have enough billions in sort of, in, in investment, and the other parts of the technology platform are rising, right, around you, you have the ability to actually deal with this, right?~
~And so, I mean, look, there are there are quantum computers where and I know this sounds nuts, right? But, like, the, the qubit is basically an atom, right? And we're doing something interesting with either the state of the nucleus or the way the electrons around the atom. And when I sell you a quantum computer that is nine qubits, I'm selling you in a three by three array of atoms.~
~How do you get the atoms there? How do you control - Yeah, how do you get them there? Yeah. I'm the wrong guy, right? Okay. I'm not. And there were earlier versions of this where the atoms. So cold atoms is a sort of a flavor of quantum computer right? And so there was a generation of quantum computers where the atoms were basically being held in space by laser beams, because at the scale of an atom, the photons basically bouncing off of you are enough to nudge you a little bit because photons have effective mass and they have effective momentum.~
~And so you have atoms suspended in space, that's very delicate, and then you, but they're close enough to each other that they can actually entangle. The newer generations are sort of more like ice cubes in an ice cube tray right? They are atoms in little architected buckets of sort of semiconductor stuff.~
~Imagine the amount of engineering necessary to make that - Yeah. Make that stuff work. Semiconducting technology has gotten to the point. I mean, transistors are, a few nanometers across, right, at this point. We've gotten to the point where we can control material reality, physical reality because of modern semiconducting at the scale, and we can deliver this stuff at reasonable economics that we can actually control this stuff.~
~So we are getting closer to the point where we can build bigger machines. And getting a bunch of qubits to entangle, it gets harder when they get bigger because they're further away, right? And there's more things that can get in the way of them and and upset them.~
~The other thing that's true, like two more bits of nerdery, all right? Only two? Oh, I will start with two. When you get them to entangle, because it is, they are so fragile, they generally don't stay entangled for a long period of time, right? And so, my laptop, when I open it up, my expectation is that if I go out for lunch and I come back, PowerPoint's still there.~
~It didn't just sort of melt. Right? Quantum computers aren't like that if you're not gonna go out for lunch, right? And the, the fancy word is that the quantum state de-coheres. When they're coherent, they're entangled. When they are not coherent, they are distant. They go back home. They go back home, right?~
~How long is that period - Yeah. For the Righetti computer? About 30 millionths of a second. Oh, well, there's no lunch. There's no lunch, right? There's no lunch. And so you - That's not happening. You arrange a bunch of computations, you get them stored in a quantum storage thing, you poke it, you figure out what they are, you set it up as the beginning of the next part of the problem, and you run some more.~
~It's like one of the reasons why you need a supercomputer on the side, right? So that engineering is cr - that engineering is crazy hard. And you know what's so interesting is people are, like, so worried that artificial intelligence is gonna take jobs, right? And here you are - Yeah. To. Yeah. I mean, that's a conversation.~
~There's quantum machine learning. Right. I mean, there are people doing. Some of the big algorithms in that space are big and hard. There are people trying to do quantum, quantum ML. I mean, and that's still crazy early. You sort of need a. I mean, to really operationalize that, you need a quantum computer, but, like, we're getting close to where you could buy quantum computers.~
~The other thing that people do, the other bit of nerdery, so one is, like, oh, it lasts for, like, 100 microseconds, right? Okay. The other is that because they are also so sensitive, sometimes they are they are, even when they are coherent, they are dis - they are disturbed or disrupted. They get noise, right?~
~And even regular computers have to deal with noise, right? When you're, like, when your transistors are 50 atoms across, right? When you're, when a bit of storage on your memory stick your terabyte memory stick is, a few, is a few hundred electrons on a capacitor, right?~
~You have error correction and error protection. And the dumbest, simplest way to do that, it's like from the early days is every time you store something back in memory, right, imagine you have eight bits, right? You add one more bit, it's called a parody bit. And the rule you have is that I will only ever have an odd number of ones in a number that I store in my memory.~
~And so if the, if it's even, I add a one, and if it's odd, I go on, okay, and I have it as zero and I store it back. And what that means is that when you read it back out, you can look at it. And if you ever see a number that has an even number of ones in it, ah, it's wrong. I should do something about this.~
~All right? That's error detection. It turns out the math is nice. If you add a few more bits, you can correct certain kinds of errors, right? So if you have eight bit numbers, you add. I think if you add three bits, you can detect any double error, and you can correct any single error, right? So there's all kinds of stuff.~
~Quantum computing, quantum computers, qubits are so delicate and so sensitive to noise and they last for such a short amount of time that one of the giant fields of study is adding error correction on top of it. Well, the problem is that qubits are expensive, right? And some of the error correcting schemes are gigantically bigger than the computing part itself, right?~
~So it's like, Here's my thousand qubits, here's my 17,000 error correcting, sorts of things. So one of the, one of the other things going on, it's like we're making progress on the basic qubits, we're making progress on the systems. And this, by the way this, this machine is also a partnership with HPE, with HP Enterprises, which is gonna help us do the integration across the conventional supercomputer.~
~And they're all over this because they see like, oh, I mean, they're a systems integrator, right? What's like a, a next great avenue of systems to integrate. And they're like, oh, quantum, I'm in. We're, we'll be a, we'll be part of your proposal. We'll help you out. Like, do we all know how this is gonna get integrated?~
~No. Maybe not. Yeah. So what? We will figure that, we will all figure this stuff out. So that's why having a starter quantum computer is great. It's big enough for where we can like, we can actually run some computations, we can learn some things. We have to figure out how we build the infrastructure to do it, but we also build the plumbing to the supercomputer next door.~
~And so everybody's like just crazy excited. And this thing then becomes a resource that is available on PSC, which is a nationally available resource. So what does that mean? I knock on your door? You can. There is a proposal submission and review mechanism at the National Science Foundation.~
~I mean, one of the things that's great about the NSF ecosystem is you can write a proposal that says, "I want to do this science. I want to do this computational thing. I think I need this kind of resources." And then there's a team inside, inside NSF and the supercomputer centers that reviews that.~
~And then if you pass the bar, you get an allocation of resources like how many GPUs or how many core hours on, on parallel computing or how much memory or other kinds of things. And you might not, you might get it on a machine far away. I mean, you might, you might live in Pittsburgh, you might get an allocation in Champaign-Urbana, right?~
~Or in San Diego or in some of the other smaller supercomputer centers. So they manage this ecosystem at a national level. Now there's a quantum machine, right, in the ecosystem. You can write a proposal, right, in, and if it's good enough and interesting enough, you'll get an allocation of, I don't know, time, probably you get a number of hours or something, right, on the machine.~
~I'm not sure how that's gonna work yet. So that levels up the country in a kind of an interesting way. And so from like a, a global geopolitical, - Yeah. Where are we in the United States in terms of - we are, well, we're, I mean, it's it's a national priority. It's been a national priority in a couple of presidential administrations, right?~
~Which is a big deal. So it's a priority for NSF. It's a priority for Department of Energy and Department of War, right? It's a, it's now a str - a nat- national strategic asset. The Chinese are investing in this, aggressively. I mean, who are our economic competitors? China and Europe, right?~
~Right. Europe is all over this. They've had the sort of the fund - the, the foundational science for, for a long time. - Do you feel like we're- We're in good, we're in a good place. In a good place, yeah. We're in a p - we're not. I mean, it's AI. I mean, I think we, we sort of though we were - But then we're not.~
~Incontrovertedly out in front and Deep Seak and a few other shots across the bow - Right. Said, - I don't think so. Maybe not as much in front as you though you were. Right. We are, we have great quantum computers. We have great scientists. We have some rising companies. Oh, I was gonna, I was gonna mention the story of the com - the, the quantum company.~
~So this is a company called Riggetti, right? It started by a guy named Chad Riggetti, who sounds like an Italian surfer, right? But he's a quantum physicist. Or a sports car driver. Or a sports car driver. Yeah. The total F1. Yeah, that's an F1, that's an F1 driver name. That's a total F1 driver name.~
~And Chad is a brilliant physicist. He's working on Josephs Injunctions, which is a technology IBM has been working on for a zillion years. There was a brief period of time many years ago that IBM tried to build a mainframe computer entirely in Josephs Injunction Technologies. So they were gonna build a, if you know what an IBM 360, right?~
~I remember that. And so these are the things that are today called the Z series machines that are still mainstays. IBM makes a boatload of money because every bank and every insurance company - They need it. On the planet - Right. Has a basement full of Z series mainframes running. Yeah, that's true.~
~God only knows how old COBOL code. Okay. And they make a lot of money now if you use. And they make a lot of money, but they are also crazy reliable. I mean, there are machines with like - Yeah. With five nines, right? Yeah. 99.99%. Like, you can go in with a shotgun and just like start shooting at a Z series and it just shrugs, right?~
~Yeah. And so they - It doesn't matter how big they are. It doesn't matter how big they are. Those, those things are still, it's a good business for IBM. There was a period of time when they were trying to make one of their, one of their mainframes go faster and better and stuff and make it out of Josephs and Junction.~
~And so it was Josephs Injunctions in Liquid Helium, all of this other kinds of stuff. It, that too early, right? It, I mean, they were able to successfully make the machines. They couldn't make them production worthy, right? They couldn't make them where like I could ship it to you and you could turn it on and you could plug it in.~
~And like the whole point of the IBM, mainframes were like, they were incredibly reliable. They just showed up and they, you plugged it in and you plumbed it and wham, off you went. They couldn't make it work, but they've had, they've been sitting on top of vast amounts of like really great Josephs and Junction sort of innovation technology.~
~So Righetti is working on this stuff and at some point, I guess he wakes up and says, "I think I can do this better." Right? And so he leaves IBM and decides he wants to start a company. He goes to Y Combinator and he gets adopted, right? He gets accepted into one of the Y Combinator classes and he basically, pitches a business plan and then everyone goes, "Oh, like that could actually work."~
~And it gets like a 25 million series A from Andreessen Horowitz. And when was this? 10 years ago. Yeah. 13, 14. I was just like looking - be a while, right? I was just looking at the webpage, to get - It's been a while. To get, to get this. I think they might have gone, I think they might have done a SPAC, right?~
~Oh, okay. Kinda sorta IPO-ish. Ba- backdoor IPO-ish, because I think they've got like a real market cap. I think we're a couple, several billion dollar kind of company. But they're now, like, you can call the sales guy, right? They'll sell you, the starter machine or a range of sort of larger machines.~
~They'll, they have so-called full stack, they have all, they have the software, right? And and all that stuff. And they're completely jazzed to be working with us because basically the Pittsburgh Supercomputer Center is a great national resource. It gets them plugged into basically the sort of like the national science network that the NSF, kind of funds.~
~And it lets them sort of work with an expert partner on like, "Well, if you need to plug this into a conventional supercomputer by working with an integrator like, oh, like HP Enterprises, like, how's that go?" Like what do you do? So everything about this is just like fun, cool, crazy, good. It's just a magic cover, right?~
~That I'm - Yeah. Yeah. I'm, it's a whole new world. Yeah. And - So it's gonna be fun. I, so this means a lot for Pittsburgh. Oh yeah. This is a nice, this is - This really means a lot for Pittsburgh. And I'm not, and there is no puffery that I'm saying about this. And so what does that mean? Full wow for Pittsburgh.~
~Right? Yeah. And so what does that mean for the things we should be doing about like shouting this to the world? Oh, we're gonna make some noise. I mean, we, I mean, just the Pittsburgh Supercomputing Center which announced it. Right. And then I think CMU announced it. We, we announced at Pitt, we jointly announced these kinds of things.~
~We've been pushing this stuff out. Righty told us that like the amount of like inquiries and sort of like press inquiries and I don't know if they got sales inquiries, they probably did, but like they said when it was announced at the Pittsburgh Supercomputer Computer Center, part of the NSF Network of Supercomputer Centers had gotten a grant to integrate a Riggetti supercomputer.~
~They said like they popped in a really big way in sort of - Wow. In sort of media attention. Right? So it's like - So what do you think that means for people that say, "I wanna plant a flag closer?" Oh, I mean, we're always in favor of more people showing up locally. I mean, you and I - But I mean, are there some strategies that we should be doing?~
~Well, I, part of the, part of what worked here is we went with one of our strengths, which was supercomputing. Right. Right? And then we went, we did a big swing, right? And so we said, like, "Does any supercomputer center have a quantum machine? No. Should we write a proposal to get a quantum machine?" Sure.~
~Sure. Yeah. Right? Is this kind of nuts? Yeah. Right. Right. But, sometimes back 40 years ago, having Ralph and Michael write a proposal for 50 million dollars to buy a supercomputer when no universities had a cray - I know. Was similarly nuts. Rigt? Right. And so sometimes, we tell everybody, you miss every shot on goal you didn't take, right?~
~So sometimes you just, you put something big and wild together and then everybody looks at it and goes, "That's nuts." Okay. "We can't not do that." Right. That's the good kind of nuts." Right? So this was that, that good kind of, "Yeah, we should let those guys, like, let's do this.~
~Let's see what, let's see what happens." And the fact that there are now starter quantum machines - Right. That don't cost 50 million bucks out of the gate, I mean, it, it is an unfortunate, you know this. I mean, it is an unfortunate truth that if you are serious about doing AI, a unit of one from Nvidia starts at about 50 million bucks.~
~Right? Yeah. So do you think Regatti what do you think IBM is saying about Regatti? I don't even wanna think about it. I have many. IBM's got their own, IBM, IBM has their own very robust quantum effort, right? So I mean, who's got game in quantum? Google's got game in quantum. Microsoft has game in quantum.~
~IBM has game in quantum. Yeah. Righty has game in quantum. And then there's a lot of smaller companies that are earlier in the in the kind of corporate formation, side of the world. They're not public. They don't have mil - thousands of employees. Right. They're small. But, they've pulled in several hundred million of venture.~
~They've built some machines. They've demonstrated viability. They have some early adopter customers who basically say, "I will buy your starter machine because I want to get to know your technology and your software stack and stuff. And then you're gonna deliver the bigger one in a couple of years," right?~
~Right. Right? Right. And every single one of these little companies is like, "Yeah, we are, this is all just hard engineering now." And we have moved from, "I wonder if I can actually do this and make a business to execute, right? I have to execute on some really tough engineering problems." And it's like, "All right, will they all succeed?"~
~No. Welcome to capitalism. So what. And so for people listening, I want them to understand, like, how many humans work for the, the supercomputer center? Oh I think it's about - 100? Yeah. I think it's like 75, 75 to 100. 75 to 100, right? Yeah. Yeah. And so - I mean, it's not a small operation. No. It is.~
~So there are. And there are. I mean, there are a couple of categories, right? There are the folks that are that, that are in the data center, sort of like managing physical machines and that stuff. But then there are also a ton of basically disciplinary experts, right, who are available to partner with people to help them do applications.~
~So, we have people who are good at life science, you know kinds of things. And like, Pitt has lots of proposals jointly with PSC, mostly in life science - Okay. Related kinds of areas. Carnegie Mellon has people doing that more in STEM side kinds of things, although, we're. And they also have zillions of partners outside, right, doing things.~
~But like, there are people there who can help you do AI, there are people there who can help you do life science. There, there are people there who can help you do molecular dynamics, right? There are people there who can help you do ph- physics-y things. There are people there who can help you do astronomy, right?~
~And all, all that kinds of stuff. So we have both the guys in the back office, if you will, making sure the machines are up and ready, and who, like, build the software systems - Right. That make it easy to do things. And, the care and feeding on supercomputers is complicated. I mean, it's not like opening the lap on my Mac or my Windows thing.~
~I mean, they have crazy complicated ecosystems. They have crazy complicated software stacks. And just like minimizing the barriers to entry right is the job of probably 50 people over there. But they're, they partner at scale. How much fun are you having? I'm having so much fun, right?~
~This - Yeah. I mean, I've known you for a while. Yeah. And listening to you now - Oh, this is. I'm like, you're amped up. This is so freaking. I mean, look, I'm a computer engineer, right? You' kidding? I didn't know that. Yeah, no. I mean, and I mean, you know this. I was joking. I mean, I spent - I was joking, Rob. All right.~
~I spent 25 years in I spent 25 years in computer engineering at the boutique engineering school next to Pitt. - I've heard of that place. Yeah. Yeah. I hear it's okay. And and then I got - What I like about you is you like morph. I, and then - You're a morph guy. And then I went - Right?~
~And then I went sideways. I got poached by the - You went to Chicago. University of Illinois at Urbana-Champaign which is - Oh, which is not Chicago. No. Sorry. It's - I always do the wrong thing. So your mental model is Stanford in the middle of a cornfield. Right. Right? And so it is a remarkable - Right.~
~Remar - and I was the head of computer science, which is a - Right. Gigantically big - kind of enterprise. And you got bored. And then I got poached back, right? And the VPR job opened up at Pitt. And I though the opportunity to come back to Pittsburgh was really good.~
~But I knew more about one of the schools here than the other. And I asked around and at everybody I talked to a bunch of university presidents and a bunch of deans, and they all said, "Oh, you gotta interview for this job." Pitt's making a lot of smart moves lately. There's like a lot of really - Yeah.~
~Like uptrending momentum. They're doing a lot of big, cool things there. And they were right. And they were right. And you've been there now for - Nine years. I know. I though so. Yeah. Nine years. Yeah, nine years. Which is 72 in human years. It's totally.~
~Okay. You're having way too much fun. People are betting on quantum. You have an amazing team of people. There's a lot of incredible research that we want the world to know why research is so important. Yeah. We want to see quantum in our lifetime, right? Yeah. Yep. We wanna see what it's gonna change and look into.~
~I think you really, I think you really are. I think, we can take a bet on fusion, right? I mean, fusion - Right. I mean, I'm, I, there's like really great advancements there but, right? The quantum stuff feels like it's gotten over the hump where it really is just gruesomely difficult engineering at scale.~
~And some subset of the corporate players in this space, either the bigs or the startups, is gonna figure it out. Well, they are. And their language is already there. Yeah. Even if they don't understand what quantum is. Yeah. And even if they don't understand what AI is - Yeah. There is a language that everyone's saying, "You know what?~
~Quantum's gotta be in our strategic plan. Yep. It's gotta be part of who we are." Yep. So anyone, any wrap up that you wanna say? I'm having an incredible amount of fun in my job. And, - Well, we see that. Yeah. And, - We hear that. You and I both know Pittsburgh is underappreciated - Yeah.~
~In nationally, right? Globally. Locally. Locally. Yeah. This is. Yeah. Yeah. Yeah. This is one of those things that causes people to look over and go, "Whoa, really? You guys got a quantum machine? Like, this is one of those, like, oh, really?" Yeah, that's what I said. Like, you guys are spending a quarter of a billion dollars to start Bioforge, sort of next gen MRNA, mRNA, factory.~
~Like, oh, wow. Like, we got lots of things like that, right? So, we're just gonna kinda keep trying to do these big swings. So what's your advice to people now thinking about that are in tech and they're either, where should they be thinking? Because it's not necessarily a market sector.~
~No. I mean, I one of my, we're in a weird, we're in a weird place right now where AI is sort of sucking all the oxygen out of the room. Yeah. Which I think is. I mean, I'm glad there's investment. I'm glad there's, there's work in this space. I mean, my last startup was a, was an enterprise AI company.~
~But I think I just saw, like, 93%. I think this was, like, Scott Galloway's, - Oh, I love Scott. Podcast. Yeah. Yeah. Right. Substack, go look, right? It's like 93% of all sort of, like, large scale venture activity is in AI. Yeah. - I've read that. Okay. Well, what happens if - Like, really?~
~What happens if that's not the only - Yeah, it's actually not the only game. Serving, right? Yeah, not the only game in town. Yeah. I mean, like, my, one of my brilliant students from Illinois went to Harvard and he and one of his buddies, another Illinois guy are doing a startup now, and they're very close to closing a.~
~And they're an they're an enterprise AI company, and they're very close to closing a $10 million Series A. And. But somebody else, I won't say who's somebody else I talked to who's in a sort of an adjacent part of the space where AI is part of their business, but they have physical stuff, bits, atoms, mechanics, logistics, supply chains, regulatory, kinds of stuff.~
~And they've said, "Wow, like, you can't get VCs to call you back because you're not a pure AI company." Yeah. It's like, all right, that gives me pause. Me too. That gives me pause. I mean, I'm happy that there's a lot invested. I have issues with all the CapEx spend and other stuff, but that's maybe.~
~I could be wrong, right? - Yeah. I mean, the only thing is we only know based on history, and everyone keeps going back to 1999 and 2000. Yeah. Which are not happy places to be looking for positively predictive - Exactly. Warm and fuzzy financial - Exactly.~
~Financial predictions. So I worry that the, I worry that the concentration, I worry that there are spectacularly cool companies a little bit choked out of the macro venture market - Yeah, I, and. Because they're not pure play AI right now. Yeah. So that's concerning. So, I mean, like, what would I be.~
~I mean, like, don't, everybody said I was nuts for my first startup, that was, that one worked well, ditto for the second one. I mean, don't take a big swing. It's okay to fail. Right? Hopefully we get through the sort of weird that is everything AI all the time now - Yeah, it is.~
~And get back to a balance kind of a thing where people can start, other biotech companies, life science companies, therapeutics companies, cancer arm, MRNA companies. But it takes a lot of academic preparation - Yeah. To be in those industries. Yeah. So we're in a weird space. Yeah. Because we're also have the dichotomy of people saying, "You don't need to go to college," right?~
~Oh I'm. Yeah. Well, I, I mean, I think Peter Thiel was straight up wrong on on that one. Taking everyone out on his boat and saying, yeah. Look, I mean, okay. I mean, for every Bill Gates who, what, made it through two years of Harvard - Right. Right, and then did something great, there are a lot of other people that actually needed a PhD in quantum mechanics to even be able to read the, the, the- They can't read the entry manual.~
~They can't - Read the, read the marketing lit, right, on the product. And I mean, I've done enough startups that you need incredibly expert people across a range of talents. And okay, if you could, if, if you were a high schooler and you were brilliant enough, all right, I'd hire you, but I hired a lot of PhDs out of CMU and Pitt, in my time.~
~Totally different. Yeah. Well, first of all, thank you for your time. Thank you for taking the big swing. Oh, thank you. Thank you for coming back to Pittsburgh. Thank you for putting Pittsburgh on a new map. And we're gonna continue to shine the light on that. Thank you for taking us down the nerd train.~
~Oh, you're welcome. I really appreciate that. I don't get to do this level of nerdery. Yeah, the nerd trainer is amazing. Nerdery very. It's like. No, I told people, "Wht are you gonna do?" This is like, "I'm going to go explain quantum mechanics to the Pittsburgh Technology Council." And they looked at me and I said, "No, this is fun.~
~This is gonna be good." This is really great. This is gonna be good. No. Yeah. This is great. And we're gonna, we're gonna continue to keep track of what's happening, shining the light, and making people more educated. I mean, because we can see the lack of education creating - Yeah. This big, have and have nots kind of world.~
~Yeah. Yeah. And I think breaking it down and letting people understand - Yeah. Solving personalized issues - Oh, yeah. That's pretty big. There's a lot of. It turns out there's a lot of workforce development opportunities - Yeah. And just like, there are people with PhDs helping you do, FinTech, right?~
~There are also people, like, wiring up networking and stuff like that, right? I mean, there's an incredible wide range, just like the, all the data centers vibe, there's an incredible wide range of skillsets required to sort of stand up these facilities. So there's a very interesting workforce development kind of angle.~
~Like we need to teach people how to program quantum computers, which is no small deal. We also need to teach people how you manage hybrid com - supercomputing - Environments. Ecosystems. So everybody, there's like all kinds of second and third order kind of positive spinoffs. People are looking to us.~
~We're like, "Oh, tell us how that works out." Yeah. It's like, "Yeah, absolutely." Well, Hail to Pit. Hail to Pit. Hail to Pit. Yeah. That was Rob Ruttenbarg. H2P. Thank you. Okay. Thank you. Thanks. Happy to be here. Yeah. Thanks Huntington Bank for always supporting us. Thanks.~
That was Rob Rutenbar, Senior Vice Chancellor for Research at the University of Pittsburgh, breaking down what a quantum computer means, not just for Pittsburgh, but for [00:17:00] the world. To recap, the Pittsburgh Supercomputing Center is getting a Rigetti quantum computer, funded by the National Science Foundation. It's a nine qubit starter machine, but even at that scale, it can explore five hundred and twelve solutions simultaneously. Scale that up to a hundred and eight qubits, and you're talking about more parallel solutions than there are atoms in the observable universe. The implications are enormous. Breaking today's cryptography, designing new drugs, simulating molecular chemistry, and optimizing problems that conventional computers simply can't touch. This is a big swing for Pittsburgh, and it puts our city at the center of a national conversation. We'll keep tracking this story. I'm Audrey Russo, and this has been the Pittsburgh Tech Council podcast. Thanks for listening.