Never-ending holiday-maker raised $2m to accelerate Australia’s quantum industry

Felix Thomsen was only meant to spend a year in Australia, but a decade later he has founded a company to accelerate the development of quantum computers.

Iceberg Quantum is designing fault tolerant chips to help make practical quantum computers a reality.

Felix Thomsen was only meant to spend a year in Australia. He convinced his parents to let him leave his home in Germany to spend his second last year at high school in Queensland.

But, the Sunshine Coast proved too irresistible to leave. And thankfully he didn’t.

More than 10 years later, Mr Thomsen is still here, has completed a PhD at the University of Sydney and has raised $2m — in a round led by Blackbird Ventures — to help the world’s biggest tech companies build fault tolerant quantum computers.

“It was initially just meant to be like a year abroad. It’s quite common in Germany to take Year 11 off and go to another country for a year, and I went to the Sunshine Coast in Queensland.

“The school was called Sunshine Beach State High School, and it was just like this long holiday and I didn’t want it to end. I somehow convinced my parents to let me stay and finish high school and then continue on to uni in Australia. And I’ve been here ever since.”

Iceberg founders (from left) Sam Smith, Larry Cohen and Felix Thomsen say they are helping tech companies develop blueprints for fault tolerant quantum computers.

He has now founded Iceberg Quantum with two of his uni mates, Larry Cohen and Sam Smith, to design fault tolerant quantum computers. The trio spent their PhD under the supervision of Stephen Bartlett, who is known as a world leader in quantum error correction, and this is what Iceberg now specialises in.

It has also partnered with PsiQuantum — the US company the federal and Queensland governments have invested $1bn in to build Australia’s first practical quantum computer. Although, Queensland’s new LNP government is reviewing the investment.

Dr Thomsen, 29, said Iceberg’s goal is to become the ARM of quantum computing. ARM has designed powerful chips, similar to the ones used in mobile phones for laptops, allowing them to perform complex tasks efficiently, which also extends battery life from a few hours to all day.

Iceberg Quantum co-founder Felix Thomsen.

ARM licenses its designs to hardware manufacturers and tech companies. Microsoft used ARM-style chips in its Copilot+ PCs, it’s first computers built for the artificial intelligence era.

“At the core of what we do is, essentially, we define layouts for how to connect up the components,” Dr Thomsen said.

“And so in the classical space, there’s companies like ARM, you know that essentially, you know, design and license out architectures for classical computers.

“At the core is just like you have a blueprint of how you connect up transistors, and we have a similar thing for how to connect up the qubits. It’s a bit more complicated in quantum because you need protocols for detecting and correcting errors, kind of as they occur.”

Iceberg’s partnership with PsiQuantum aims to develop fault tolerant architecture’s to solve the biggest problem with quantum computing, how to make them less error-prone.

Microsoft’s vice president of advanced quantum development Krysta Svore.

Quantum computers already exist, albeit on a small scale and unreliable form. The difference between quantum and classical computing is quantum uses qubits rather than bits. Bits can either be 1 or 0 to process information, while quantum could be both at the same time — like a coin being both heads and tails spinning in the air before it lands — a process known as superposition, and is incredibly powerful.

Quantum can process information in a fraction of the time of classical computing, in some cases saving thousands, maybe millions of years.

But, qubits are, as Microsoft’s vice president of advanced quantum development Krysta Svore says, “noisy”. She says they are like 1000 spinning tops, and the task is to keep them spinning for a month, while the room moves around a bit. “It’s a hard task”.

“Where we are today? We think of one in a thousand operations fail, but as we scale up, we need only one every billion operations to fail. That’s practical compute,” Dr Svore says.

Microsoft's Majorana 1 chip, which will help pave the away for practical quantum computers.

Microsoft has developed a quantum computer with 25 logical qubits, and plans to double this to 50 by the end of the year. If it gets to 50 logical qubits, it can start to perform tasks you can’t do classically. It has invested more than $US1bn in quantum and has created a new chip which leverages a new state of matter and that could underpin quantum computers.

It uses what’s called a topological superconductor — a type of material which can observe and control particles to produce more reliable and scalable qubits, which it says are the building blocks for quantum computers.

To start doing things on an industrial scale it needs to get to 150 logical qubits and beyond. And, if it wants to start changing chemistry, it needs at least 1000 logical qubits — a process which could take years.

Blackbird partner Michael Tolo.

Iceberg’s approach involves using LDPC codes — a type of error correction code — to reduce the hardware overhead required for fault tolerance by “more than an order of magnitude”. Dr Thomsen said this enables quantum hardware companies to achieve practical fault tolerance faster, while reducing the cost to build and operate useful quantum computers, and “unlock more valuable applications by increasing computational power”.

Blackbird Ventures partner Michael Tolo says this could significantly accelerate the timeline to deliver a practical quantum computer.

“We are fast-approaching the useful quantum computing era and are proud to see Australians leading the charge,” Mr Tolo said.

“The technical progress that Felix, Larry and Sam have made over the last six months shows us that their approach could accelerate the timeline to useful quantum computing and push these systems to perform more valuable tasks, sooner.

“We had the privilege of watching the team grow since participating in our Foundry program last year, before the company had even been incorporated. We invest behind unique insights and velocity of progress, and this team has both in spades.”

PsiQuantum chief architect Terry Rudolph said: “Australia has long been a global leader in fault-tolerant quantum architectures, setting the direction for the field”.

“Its next generation of talent will play a pivotal role in shaping the future of quantum computing,” Professor Rudolph said.

Originally published as Never-ending holiday-maker raised $2m to accelerate Australia’s quantum industry