How tinkering with electronic circuits led John Martinis to a Nobel Prize in Physics
…And then he smiled.
The audience welcoming John Martinis, this year’s Nobel Prize co-winner in Physics – mainly students and postdocs at a large auditorium at the Tel Aviv University – fell silent. After all, it wasn’t everyday they’d be listening to a Nobel laureate. This talk, on 30 November 2025, was one of many he’d held around the globe, on a Nobel talk marathon following the announcement on 7 October that he, alongside Michel Devoret, Chief Scientist for Quantum Hardware at Google Quantum AI, and John Clarke, Professor of Physics Emeritus at UC Berkeley, got the prestigious award, for their trailblazing work in quantum back in 1985.
A Nobel marathon that is culminating this week, with the trio jointly receiving the prize in Stockholm on 10 December, and the whole world tuning in.
Martinis came to Israel on his way to Sweden to make an important announcement: the company he co-founded, Qolab, had just deployed its first chip outside of Qolab’s home base in Madison, WI, at the Israeli Quantum Computing Center (IQCC). This lab, on the grounds of Tel Aviv University, is operated by Quantum Machines, and the three organizations are working in close synergy helping Qolab build a quantum computer, with Quantum Machines supplying state of the art control systems.
“It’s obviously a great honor to be awarded the Nobel Prize,” says Martinis, a tall, grey-haired and seemingly perpetually warmly smiling researcher. “But the biggest honor of all was to actually do the experiment, publish the paper, communicate it, and also to develop the field over the years.”
He’s incredibly excited, he adds, that quantum computing has become such a large field with so many teams working on building this emerging technology that mimics nature. “I really like that quantum computing has developed into something that’s, frankly, practical – once we are able to build the machine and solve practical problems. Moving it from basic science to practical applications has been fantastic to see over 40 years or so.”
Early curiosity and the path to quantum physics
For Martinis, now 67, the fascination with science began when he was in secondary school. More specifically, first came the fascination with electronics and building things in his parents’ garage. “My father was a fireman, and when he was at home, he was always building various projects for a dune buggy or for a cabin in the desert,” chuckles Martinis. “He was very smart and would have been a great engineer. He had this understanding of things in a certain intuitive way, which he transferred to me. And when I took physics in high school, I could put the math behind everything that I already intuitively understood. That’s what got me so interested in physics.”
As a young kid, Martinis realized that math made things work, turning ideas into practice. One day, he recalls, he built his own electronic circuit for the first time, having read a book about circuit design and applying mathematical equations to make it work. And it did.
This passion for electronics, understanding how the world ticks and building machinery to assist us with that quest, has taken Martinis on a career spanning academia and industry, from working at tech giants to kicking off startups. Having finished his undergrad at UC Berkeley, he decided to stay there during his graduate studies, too, under PhD advisor John Clarke – in part because Clarke was working on electronic devices. “A big part was also that John was looking at quantum effects in these electronic devices, and I thought it was really interesting to put the excitement of quantum with kind of my hobby of doing electronics,” says Martinis. “And really enjoyed the way John Clarke built systems, built practical, SQUID measurement systems out of the research he was doing.”
Proving quantum mechanics in the macroscopic world
While a postdoc in late 1980s, Martinis began attending more and more conferences, and at some point caught whiff of the idea of macroscopic quantum tunneling. Researchers were working on some initial experiments at the time, but what excited him most was the interesting theory behind them. He talked to Clarke about conducting an experiment, too, and together with Michel Devoret, the trio decided to give it a go. Their work on macroscopic quantum tunneling and quantized energy levels in Josephson junctions proved for the first time that macroscopic electrical circuits behave quantum mechanically – and ultimately led to them to receiving this year’s Nobel Prize.
After doing his postdoctoral work at NIST Boulder, Martinis got tenure at UC Santa Barbara in 1992, and his superconducting qubit group became one of the most highly regarded in the world. Fast-forward to 2012, the year when he and his team published an influential theory paper on surface code error correction. It was that work that helped lay out a roadmap toward developing a fault-tolerant, scalable quantum processor.
Quantum error correction is still a crucial area of research, a challenge en route to a fully functional quantum computer. Qubits are incredibly fragile, and to keep them in a quantum state researchers operate at millikelvin temperatures and use elaborate shielding to suppress disturbances, aka “noise” that leaks in from temperature drift, electrical fluctuations, stray radiation, and countless environmental factors. But noise always finds a way in. Today’s leading approach is to engineer hardware that can detect and correct these errors in real time before the final measurement is taken.
Together with another seminal paper in 2014, “Superconducting quantum circuits at the surface code threshold for fault tolerance,” Martinis’s efforts laid the foundation of the more recent surface code milestones, including the development of Google Quantum’s Willow, a 105-qubit superconducting processor announced December 2024, where scientists implemented a surface-code quantum-error-correction scheme.
While in 2014 it was clear to Martinis what machine he would have to build – a surface code error corrected quantum computer – resources in industry and academia often differ. So that same year, when the opportunity arose for him to move to Google, “it seemed like an ideal situation,” he says. At Google, he reckoned, he’d have a nice combination of resources and an opportunity to work on hard engineering problems all the while producing amazing science.
“I brought the postdocs and students over, and we were able to bring up the technology pretty nicely. And that culminated in the quantum supremacy experiment in 2019,” recalls Martinis.
Shaping the future of quantum computing
That work, published in Nature, generated a media frenzy and helped to make the term ‘quantum’ a near-household name. Unfortunately, Martinis says, shortly after the paper’s publication Google decided to manage the quantum program differently. “I just kind of didn’t fit in at that point,” he adds. He left the tech giant in March 2020 – which was, in part, also because he “had this bigger vision of what a quantum computer should be,” he adds. “After leaving Google, I really thought carefully about what we needed to do and how we had to really radically change the way you would build this superconducting quantum computer. That’s what I’ve been fortunate enough to work on with our company.”
His company, a startup Martinis co-founded in 2020 with another former Googler Alan Ho in Palo Alto, is called Qolab for a reason – it’s short for ‘quantum collaboration.’ The idea? To build a quantum chip with better quality qubits than what’s currently out there, and to do so by collaborating with the best experts in the field.
Qolab aims to push this frontier by designing devices with inherently lower noise and more stable qubits, reducing the burden on error-correction schemes and making scalable quantum computing more practical. Researchers working on error correction, including those at Google, have “done very well showing that the error correction works,” says Martinis. “But I think they have to make a little bit more progress. And to get to that, it’s going to help to fabricate the devices better, which is what we’re working on – to be able to scale up to millions of qubits.”
‘Better’ means building the qubits more reliably, with better yield and better performance. “Because right now, even at a hundred qubits, the systems are still a little bit too artisanal, let’s put it that way, to think about scaling up to a million qubits,” explains Martinis. “We’re working with applied materials to make these qubits in a much more modern way, using deposition and etch, the way people build billions of transistors. And we have a new architecture, based on wafer scale integration.”
So instead of using the familiar superconducting approach with cryostats, multiple wirings and microwave components, with a chip buried inside, Martinis’s idea is to put all of that on a single chip, a single wafer. “We think that will scale better and allow us to scale up to a million qubits, much, much better,” he says.
Being a small startup, Qolab isn’t at it alone. Just like for any quantum computer, a control system is a critical component. That’s where Quantum Machines comes in, with the most modern fabrication, architecture, and electronics based on full digital synthesis – exactly “what you’re going to want to do to scale up,” says Martinis. “We’re a tester of Quantum Machines’ electronics and we’ve tested them very carefully – there’s very good interaction. We’re also working with them on various noise issues so that it’ll measure our qubits better. It’s an extremely tight and good collaboration.”
What’s next for quantum? Martinis hesitates slightly, not wanting to give predictions for the entire field. For his company, though, he’s game, radiating optimism: “I hope in the next months and years, we can start showing that if you build qubits in this different way, they’ll be much better and more manufacturable.
“And I hope that in a few years, we can have people re-envision what these quantum computers will look like and very quickly be able to scale that up, because we’re working with people who know how to manufacture things properly.”
