Exploring Quantum Frontiers with Michael J Biercuk, CEO & Founder, Q-CTRL
Please give us a little introduction about your current role and what you do:
I’m a Professor of Quantum Physics and Quantum Technology and have built Q-CTRL as a world leader in quantum infrastructure software. We’re focused on making quantum technology useful through our expertise in quantum control engineering.
Do you see any breakthroughs in error correction this year?
It’s an incredibly exciting time for the development of Quantum Error Correction and we do expect continued demonstrations that make clear that QEC has a real useful future ahead.
We also think it’s essential to help the community understand what new capabilities - “breakthrough” or not - actually enable. To this end, we’ve focused on the practical challenges of enacting QEC in a way that gives real benefits, and helping end-users interpret hardware roadmaps. For instance, we recently introduced the new concept of “QEC advantage” because most readers who aren’t experts won’t understand that even with new capabilities in QEC emerging. QEC may actually be an inferior choice relative to other error reduction strategies, and this may persist for many years because of the relatively weak performance of demonstrations thus far and the incredible overhead required.
We are very keen to help the community develop an informed view of how to build a quantum computing implementation roadmap in the face of complex and rapidly changing technical advances.
Where do you stand on the future relationship between NISQ and fault-tolerant quantum computing?
Our core goal is to help build machines that are “fault-tolerant enough” for emerging applications. It remains a moving target of what “enough” may be, but we are excited both that systems are getting more capable, and applications are being identified that can deliver advantage with lower resource requirements than previously thought possible (for instance certain problems in CVRP - capacitated vehicle routing for transport). Error will always be a major bottleneck and we’re working to deliver the best possible technology to deal with it, whether it’s pure error suppression before QEC, or a hybrid QEC implementation designed to deliver maximum performance and efficiency.
In our view the coarse-grained view of “NISQ vs Fault Tolerance” is not helpful to either advancing the field or enabling end-users understand how to plan for quantum implementation roadmaps. It’s not just an oversimplification - it also actively misleads. For instance, just because QEC functions beyond breakeven (ie. it delivers a net benefit relative to the resources it consumes) it most certainly does not mean that there are no longer any errors in quantum computers. Given that, one has to realize that even beyond this point, QEC may not actually be better than alternate strategies in dealing with error! This is likely to come as a major surprise to many!
We’ve been working hard to ensure that we can help end-users parse this complex landscape and gain real insights into when and how they can benefit from quantum computing for problems that matter to them.
How many qubits do you see us reaching in the next year to 18 months? Will they make a difference?
We’ve been excited by industry roadmaps and deployments of larger and larger machines. The key question is what’s possible for an end-user to achieve with these larger systems.
Often there’s been a large gap between what an internal R&D team can achieve and what is made publicly available. We’ve focused – through our performance-management infrastructure software – on pushing hardware to the limits in a manner that delivers any end-user with a user experience equivalent to the best an internal team can achieve.
This is exactly what we’ve achieved with our native integration into IBM Quantum Services, giving a zeroconfig solution so any end-user can push the hardware to intrinsic limits. And it really works - in our tests with these publicly available tools we’re running algorithms successfully with full devices at utility scale.
All of this capability has been fully automated and abstracted so an end user doesn’t need to be an expert in hardware characterization, error sources, or the like in order to take advantage of the “latent” performance inside the hardware. Everything boils down to a single command so any user can take advantage of quantum computing for their problems.
And we’ll continue pushing in this direction. Quantum infrastructure software allows abstraction, “virtualizing” the underlying hardware in such a way that an end user doesn’t even need to worry about quantum circuits! We already released a fully configured QAOA solver in Fire Opal that takes care of all aspects of full QAOA implementation, and we’ll be releasing further abstractions as we focus on making these machines more accessible to teams with conventional IT skills (rather than quantum physics skills).
Michael will be speaking alongside QuEra Computing, IBM, Quantum Computing Inc and Pasqal providing insights and answers to a key question ‘where do the major platform and processor providers now stand?’ at 11:45am on April 25. Download the full agenda here today.
Where do you stand on the current talent shortage in quantum? What more do you think can be done?
There’s certainly growing demand for expert practitioners capable of slotting directly into active quantum R&D programs.
Much of the response at the governmental level has been to prioritize PhD funding. We support this and will continue to vocally endorse investment into basic research programs to produce skilled experts.
We also believe it’s essential to help experts from adjacent fields - such as computer science or engineering - enter the field and contribute their skills without the need to also have a PhD in quantum physics. Our view is that this class of employees is often overlooked in policy discussions, and so we set out to provide a solution on our own.
Black Opal is a world-leading edtech solution to help anyone learn quantum computing without any formal background. It’s mobile friendly, modular, and highly interactive to help build intuition about complex topics - no math prerequisites. It was really built for exactly these “adjacent” experts - folks who want to enter the field, or even start advanced training but really just don’t know where to start. We’re proud that it’s become the default skills training program in the UK via the NQCC, and that it’s used globally by Fortune 500 firms, major consultancies, and even Universities as they help non-physicists learn quantum computing.
How do you see quantum working with AI and machine learning?
We build AI-powered quantum infrastructure software, so we’re very excited about the convergence! Our focus is on how AI and ML can enable us right now to better control and manipulate quantum hardware. We’ve built all of our own agents (no chatGPT wrappers here) and shown that we can automate tasks that were previously highly manual and therefore limited to a narrow range of experts. That is, embracing AI has been key to allowing us to broaden the accessibility of quantum computing to end-users who don’t have specialist knowledge of the underlying hardware.
What are you looking to showcase at our show in Washington DC this April?
We will be very excited to showcase our zeroconfig and fully automated performance management solutions that allow any user to achieve outputs at the limits of what’s possible for hardware. There’s no need to be an expert in quantum physics or even knowledgeable about the underlying hardware. Our software enables full abstraction or “virtualization” so any application gets a major boost.
As systems have grown the advantages have grown as well. With a single command it’s now possible to boost the functioning of a key application called QAOA by over 100,000 times. And with the demonstrations we published comparing against the open literature, this software indisputably makes the supported IBM systems the most powerful quantum computers available in the market.
Those boosts can bring totally new applications within reach, by simply unlocking latent performance in the hardware.
Quantum.Tech covers quantum computing, cryptography as well as sensing; what area are you most excited about and why?
We’re very much looking forward to sharing how the same approach we take to enabling quantum computing hardware with powerful performance-management software is unlocking new capabilities in quantum sensing!
We’ve focused extensively on addressing the challenges that have blocked the deployment of quantum sensing in real environments. Of course this involves careful hardware engineering led by our expert team, but it also involves learning how to trade hardware for software-based stabilization whenever possible!
Our peer-reviewed technical manuscript showed how atom interferometry could be made effectively 20X more tolerant to environmental disturbance via software ruggedization, as compared to using conventional operating approaches.
The unique capabilities we’ve built to power fielded quantum sensors have enabled us to become a core provider of quantum-assured navigation with AUKUS partners (as featured in the New York Times!). We’re very much looking forward to engaging with US Defense and Security partners as we seek to deliver truly transformational capabilities to allies.
As well as appearing on the panel discussion on platform and processors, where we stand on it today?, Michael will also be joined alongside Riverlane, Quantum Machines, Bluefors, Fermilab and Rigetti on April 26 looking at ‘Delivering high performance on-premise quantum computing capabilities. Make sure you book your place at Quantum.Tech USA here today to hear from Michael J Biercuk explore the advancements of quantum technologies.