How does a guy go from being the engineering director for Amazon Prime to serving as the CEO of a quantum computer company? It’s a classical move for Peter Chapman, the president and CEO of IonQ, which provides the firepower for Microsoft’s recently announced Azure Quantum cloud computing platform.
Quantum computing promises to address the same kinds of optimization problems that Chapman had to deal with for Amazon’s next-day deliveries, but on a grand scale. It also doesn’t hurt that Chapman previously worked for futurist Ray Kurzweil, or that he believes quantum computers provide the only path to strong, human-like artificial intelligence.
“I really like that kind of bleeding edge,” Chapman told GeekWire.
IonQ should be edgy enough for Chapman, who became the company’s top executive in May. It’s on the bleeding edge of quantum computing, with an unconventional technology that relies on the properties of trapped ytterbium ions.
When handled properly, the ions can represent multiple values simultaneously, in line with the dead-and-alive paradox made famous by Schrödinger’s cat. Those ions serve as quantum bits, or qubits, theoretically producing computational results that are impractical or impossible to produce using classical binary operations that are limited to ones and zeroes.
‘The most boring qubits out there’
It all sounds exotic, but Chapman said the trapped ions behave in line with straightforward physics.
“It’s based on atomic clocks, which have been around for a long, long time,” he said. “In terms of the qubits themselves, there’s nothing exotic about them. To be honest, they’re probably among the most boring qubits out there.”
Chapman and Chris Monroe, a physicist at the University of Maryland who co-founded IonQ in 2015 and now serves as the company’s chief scientist, said that very boringness could give them an edge over bigger companies like Google and IBM.
Those companies are following a different technological track that relies more heavily on advances in superconducting materials. They’re definitely making advances. In October, researchers at Google announced that they had achieved “quantum supremacy” for a particular kind of random sampling problem. But Monroe said it’s not clear how much farther they’ll be able to get.
“Right now they’re seeing scaling issues,” he said. “No. 1, each qubit is a little bit different. No. 2, they have to be wired up.”
Those are two issues that don’t have to be addressed with trapped ions, which are all identical and interact on the atomic scale with the aid of lasers. “We face our own problems, but they’re totally different,” Monroe said.
The challenge facing IonQ is to knit together an increasing number of qubits while maintaining control over the connections. Monroe said the company’s researchers have demonstrated fully interconnected systems in the range of 50 qubits, and are “pushing well into the 20s of qubits” for computational purposes.
“We’re very careful about scaling up,” he said. “We’re not going to throw a million qubits on the table unless we can do millions of operations.”
A quantum leap … to the Seattle area
IonQ’s progress to date has attracted attention from investors as well as from Microsoft. In October, the company said it secured $55 million in a funding round led by the Samsung Catalyst Fund and Mubadala Capital. Chapman’s former employer, Amazon, joined in the round, as did Airbus Ventures, Hewlett Packard Pathfinder, GV (a Google spin-out) and other venture funds.
The partnership with Microsoft could well mark a quantum leap for both companies. IonQ is setting aside one of its trapped-ion systems to take on quantum computing tasks that are fed into it with the aid of Microsoft’s Q# programming language, Quantum Development Kit and Azure cloud platform. As usage increases, more systems will be brought online to handle the load.
“We expect demand to be pretty high,” Monroe said.
So high, in fact, that IonQ is opening an office in the Seattle area, most likely in Bellevue, to focus on the Azure Quantum collaboration. The Maryland-based company’s workforce of about 50 employees already includes a handful in the Puget Sound region, but now that the partnership has gone public, Chapman said it just makes sense to have a higher-profile presence.
“In addition, Seattle has a deep pool of talented engineers across many industries, and IonQ believes it’ll be a tough sell to get those potential new employees to leave the Pacific Northwest,” he said in an email.
Chapman knows about that allure firsthand, thanks in part to his time at Amazon. Even now, he’s splitting his time between Maryland and North Bend, Wash.
The big question about quantum
So what’s quantum computing good for? To answer that question, Chapman turns to his experience as Amazon Prime’s director of engineering.
It was up to Chapman and his team to come up with algorithms that could figure out the fastest way to get a product from a supplier’s warehouse to the customer’s doorstep. If the customer selected, say, a USB drive, Amazon’s servers would sift through about 6 billion products to identify 14 million items in inventory.
“In that one second, we went through and looked at all 850 fulfillment centers where that product was located, for each one of those 14 million. Then, realizing it was going to take 45 minutes to get it picked, packed and put in a box, and whether it’s going to hit the 3 o’clock truck, but the 3 o’clock truck is going to be 98% full, and so maybe it won’t go until the 8 o’clock truck … we did the whole thing for all 14 million items in that one second,” Chapman said.
“The average person doesn’t have a clue that that’s what’s going on,” he said, “and that’s the problem set that I worked on.”
In-depth: What will quantum computing do for us?
Computer scientists say such optimization problems should be tailor-made for quantum computers, because they can sift through multiple permutations of a complex system far more quickly than classical computers can.
And we’re not just talking about delivering USB drives. The potential applications range from hack-proof cryptographic systems to new approaches to molecular chemistry. Microsoft and its partners are already trying to employ quantum computing to unlock the mysteries of biochemistry.
“The chemistry problem sounds a little narrow, but it is an optimization problem,” Monroe said. “And if we get good at that, of course we’re going to be looking at financial models and how to minimize those, and logistical models and how to minimize those. Everybody out there has their own problem of optimization, and they’re largely ignored because they can’t solve them.”
Chapman has an even grander view of quantum computing’s potential, thanks in part to his experience as an executive at one of Ray Kurzweil’s companies. Kurzweil, of course, has long predicted a technological singularity brought about by the increasing power of artificial intelligence.
“I have a strong interest in strong AI,” Chapman said. “I’ve had that all the way back to the very beginning. And I think you need a quantum computer to solve strong AI. So, not only do I want a quantum computer for the world, but I want one myself — to go finish that problem.”