Meet the man building the future of quantum for IBM, qubit by qubit

Follow the Hudson River an hour north of New York City, just past the Tappan Zee, and you may find yourself outside a curious crescent-shaped building. Though it’s lined entirely with windows, not one offers a peek at what’s really happening inside. For that, you’d have to follow maze-like routes into the center and somehow bypass the iris-scanning devices that stand guard at each doorway.

Then, and only then, would you lay eyes on IBM’s quantum research laboratory, housed in its T.J. Watson Research Center. It’s packed with peculiar gold instruments; fridges colder than outer space; whiteboards full of complex equations; and niche comic strips about science, cut from newspapers and taped all over.

At the center of it all is Jerry Chow, one of the quantum world’s most influential scientists.

Chow, director of IBM Quantum, has been working to uncover the technology’s mysteries for more than a decade. He has risen through the company’s ranks since joining in 2010, leading IBM to announce a milestone achievement in 2021: breaking the quantum world’s famous “100-qubit barrier.” He also spearheaded the first-ever cloud-based quantum processor in 2016, making quantum computing publicly accessible for the first time—the single biggest catalyst for experimentation in the field, Moinuddin Qureshi, a computer science professor at the Georgia Institute of Technology, told us.

“Before 2016, probably a few hundred people in the world [had] ever experimented with a quantum computer,” Qureshi, who worked as a research staff member at the T.J. Watson Research Center from 2007–2011, said. Since 2016, IBM said it has had more than 400,000 users of that platform. “To me, that has been the defining change,” Qureshi added.

Tens of thousands of people have heard Chow explain, via a 2015 TED Talk, why quantum computing matters. To paraphrase him, one reason is that no existing computer is advanced enough to explain the inner workings of the particles that make up our world. Even if someone packed a traditional computer chip with the same number of transistors as there are atoms in the galaxy, it still wouldn’t be powerful enough to understand even a molecule like caffeine, let alone the secrets of complex viruses or DNA.

A quantum computer could eventually be able to do that and much more, finding patterns and insights from data that could one day lead to pie-in-the-sky applications like early knowledge of a pandemic virus, more powerful batteries for electric vehicles, or even treatments for cancer.

But the field has to overcome several fundamental hurdles before it can achieve its imagined potential. Though Chow led IBM to its major 100-qubit breakthrough last year, Qureshi said it’s likely that a computer capable of the goals above would require a few million qubits—the equivalent of bits, or 1s and 0s, for traditional computers. That’s a task that could take years, if it’s even possible.

Chow isn’t alone in his mission. Google, Intel, Microsoft, and Amazon are all investing in their own quantum-computing efforts. Since March 2021, three quantum-computing firms have gone public, all via SPACs, ranging from $1.5 billion to $2 billion. China is also advancing in the quantum-computing race, having developed two of the world’s most powerful quantum computers. And last week, President Biden signed an executive order that seeks to “ensure continued American leadership in quantum information science and its technology applications.”

For his part, Chow is attacking the field’s challenges by simultaneously toiling away at hard science questions and productizing what IBM has built so far. By taking what’s “good enough” and putting it into the market, his team is collecting vital feedback from users that, hopefully, will accumulate into aha moments over time.

“There’s [one] piece, which is like, ‘Okay, I’m going to go head-down and make improvements,’ but there’s another piece, which is, ‘Okay, I’m going to actually make a product and have it [be] accessible and basically almost get feedback on what’s working well, what’s not working well, what kinds of algorithms are getting developed on the hardware we’re building,” Chow told Emerging Tech Brew. He added, “It’s breaking the mold of just linear research and development.”

When walking around the IBM quantum campus with Chow, it’s easy to see that his colleagues view him as central to the tech’s future. Whether they’re standing on a stepladder building a quantum computer, pushing a tank of liquid nitrogen down the hall, or tinkering with an Allen wrench, everyone seems to want face time with Chow—and he makes sure to return every wave as he completes up to 10,000 steps per day.

But once upon a time, Chow was the one trying to steal some time with a quantum expert.

The single qubit days

It all started with a knock on an unfamiliar door in 2002.

Chow, then a sophomore at Harvard, recalls he had just come across an article on a professor’s work with quantum dots. On a walk through the physics lab, he spotted an office placard bearing the name “Charles Marcus”—the same researcher featured in the article he’d read. Chow mustered up the courage to rap on the door, told Marcus he’d read his work, and asked about getting involved.

Working in Marcus’s lab became Chow’s first hands-on exposure to qubits, the secret sauce that makes quantum computing so powerful.

In traditional computing, the smallest possible unit of data is known as a “bit.” Bits can have one of two values, either 0 or 1, and they’re typically represented by electrical pulses that mean one or the other. Everything you do online, including reading this article, is made possible by strings of these 0s and 1s.

Just as bits are the foundation of traditional computing, qubits are the building blocks of quantum computing—the smallest possible unit of quantum information. Qubits are essentially subatomic particles like electrons or photons, but they’re a lot more capable than their binary counterparts. That’s because qubits have many more than just two potential states: One of their mysterious properties is their ability to represent many different combinations of 0 and 1 simultaneously.

But there’s a catch: The particles are extremely unstable, and it’s difficult to keep them in their ideal state long enough to perform the complex calculations they make possible.

Chow wanted to see what would be possible when qubits were given a longer lease on life. So in 2005, he went to study the subject as a graduate student at Yale.

At Yale, he met his match in Jay Gambetta, an Australian postdoc with a background in theoretical physics. Together, they worked on another mysterious quantum property: quantum entanglement. Chow and Gambetta wanted to prove that they could use a “gate” to couple two superconducting qubits. If they successfully entangled the particles, they would have showcased a critical building block for making quantum algorithms work. They recalled that their research eventually led, in 2008, to the first-ever demonstration of simple quantum algorithms on a superconducting quantum processor.

In New Haven, Gambetta and Chow had spent their days in the lab and their evenings drinking Pabst Blue Ribbon at the graduate-student watering hole. When Gambetta moved on from Yale, they still found a way to work together, playing Halo for hours each evening while they talked through the “what” and “why” of potential experiments.

“Jerry’s always been extremely fast at picking up the ‘why’ and being able to iterate very fast with the hardware to be able to demonstrate it, to get to the ‘how,’” Gambetta said. He added, “I remember even back at Yale, we would talk the night before…playing video games, and then he would send me results that he had done from the experiment in the morning.”

Fast-forward to 2011, and Chow, as research staff at IBM, helped recruit Gambetta for a role on the team. They were once again working together in person on a similar type of two-qubit gates as before, this time with a new approach: microwave controls instead of magnetic-field controls. On any given weeknight, Chow and Gambetta could be found in an out-of-the-way 10x12 office, surrounded by cables, network analyzers, and oscilloscopes, running experiments to make the gate work.

Their experiments turned out successful, and as the quantum buzz continued to build, Chow recalled that physicists and theorists everywhere were asking about ways to use IBM’s hardware and work together. To Chow and his team, it seemed like too many potential collaborations, too little time. So he and Gambetta began working on a solution—and less than five years later, in 2016, they put the first-ever quantum processor in the cloud. That way, anyone who passed an approval process could access quantum computing for free.

Just as quantum computing was advancing as a field, Chow himself was entering a new phase of his career.

From researcher to manager

As a research staff member, Chow was known for his passion, which took the form of whooping and clapping loudly when an experiment went right and releasing strings of expletives when it went wrong.

But starting in 2014, in his managerial role, his day-to-day began to look different: less hands-on lab time, more delegation, and a lot—a lot—more meetings.

Now, as director and IBM fellow, Chow’s morning routine starts with a 6:30am wake-up call from his three-year-old daughter. After dropping her off at preschool, he grabs a 20-ounce coffee at work, then checks in with Gambetta, now IBM fellow and VP of IBM’s quantum-computing division.

“If I can walk by Jay’s office, and he doesn’t call me [in], I know that there’s not a fire,” Chow said. “I sometimes get just past his door, and he’s like, ‘Can you come in here?’ And it’s like, ‘Okay, what do we gotta do?’”

Currently, Chow oversees a number of different labs, which are each working to solve a quantum quandary. Some are focused on solving the field’s most persistent issues, like improving coherence, or the amount of time that qubits can stay in their useful states, or issues with quantum entanglement, another core element of the technology’s power.

It took Chow time to adjust to managing people; he recalls asking his wife for advice on conflict and communication. But to some members of his team, Chow’s move from quantum researcher to tech lead seemed seamless.

“He was a very strong researcher and really key element in quantum, but now…he’s really driving the technology,” Mary Beth Rothwell, senior manager of experimental quantum devices, told us, recalling that she had no idea IBM had a quantum department until she joined the team in its early days.

“I didn’t even know they existed down there, because it was really kind of this little small entity at the end of the building…then all of a sudden, I joined, and it was like this really close-knit group that really were focused on this mission, and we really wanted to prove that this was viable,” Rothwell, who has worked on quantum at IBM since 2007, said. She added, “I’ve been with IBM a long time—over 30 years—and this is the most exciting project I’ve worked on.”

Quantum’s course

Chow’s ultimate goal is to bring quantum computing into the modern toolbelts of scientists, academics, and industry—to turn its myriad applications from potential to reality. That’s the type of world he wants his daughter to grow up in.

But for now, quantum computing’s most promising applications are largely theoretical.

And developing the right building blocks is anything but smooth sailing. Chow, his team, and the rest of the quantum world are working to solve a wide range of technical obstacles that suggest the tech’s potential transformative applications are far off.

The global quantum-computing market, worth an estimated $412 million in 2020, is projected to reach $8.6 billion in revenue—and more than $16 billion in investments—by 2027, according to the International Data Corporation (IDC). For comparison, global private investment in AI surpassed $93.5 billion in 2021.

Georgia Tech’s Qureshi said that IBM’s breakthrough 127-qubit chip produces only about 16 bytes of information. With that view, “there’s a long way to go,” he added.

Still, researchers are paving the way. At least five quantum-computing companies have publicized plans for quantum-computing hardware that’s “fully fault-tolerant,” i.e. with relatively low error rates, by 2030, according to a McKinsey report. While IBM has previously indicated it hopes to build a million-qubit machine by 2030, the company told us it does not have a firm goal for 2030 yet.

Despite the competition, Qureshi said, IBM appears to be pulling ahead in some respects, due to its software infrastructure, daily quality reports—and Chow’s work on cloud-based quantum. Qureshi said IBM has invested in building a quantum-computing community and lowering the barrier to entry by setting up software infrastructure, creating classes, and offering opportunities to high school and college students.

In terms of technical specs, IBM and Google have been competing for years, but Qureshi believes IBM is paving the way in terms of wide-ranging impact.

“In terms of technology improvement, I would say that IBM might not be the top one—it’s certainly top two,” Qureshi said. “If you just look at the quality of qubits, IBM and Google, I would say, are head-to-head. The quality of qubits is somewhat better in Google. But in terms of a broader impact…in terms of educating the community, creating the workforce, creating a line of people who can access quantum comput[ing]…that, I think, IBM is doing a fantastic job [at].”

He believes that IBM’s multidimensional approach to quantum computing could be key to its success.

“Leadership is not just about ‘how good are your qubits’—leadership is also about setting up an environment for the technology so that adoption happens,” Qureshi said.

For now, though, Chow is focused on simply building hardware as quickly as possible, even if it’s imperfect, and iterating from there—qubit by qubit.

His father, a nuclear physicist, was strict, Chow recalled—jokingly calling his son his “worst student” when Chow lost interest or fell asleep during impromptu lessons. But in Chow’s own journey to create flawless quantum particles, he’s discovered that you can still learn a lot from “good enough.”

“Throughout the journey, especially with these superconducting qubits,” Chow said, “it’s been [those] kinds of incremental steps that have really brought us such a long way.”