BlueQubit: Shaping the Future of Utilizing Powerful Quantum Computers
Quantum computers promise a new era of computing power poised to revolutionize various industries. However, operating a quantum computer remains daunting, often requiring deep expertise in low-level programming, hardware engineering, and even quantum physics.
The challenge for users of quantum computers is not only to understand and operate these complex machines but also to benchmark their capabilities and estimate what they will be able to do once they become widely available.
BlueQubit is on a mission to democratize access to quantum computing through a user-friendly platform that allows developers and researchers to harness the power of quantum hardware with a single click. Founded by Hrant Gharibyan and Hayk Tepanyan in the spring of 2022, it went through the Creative Destruction Lab startup program.
Learn more about the future of utilizing powerful quantum computers from our interview with the co-founder and CTO, Hayk Tepanyan:
Why Did You Start BlueQubit?
Since I was a kid, I have been fascinated by computers. Few people know what they’ll choose as a major for their studies, while for me, it was clear that I wanted to do something with computers. I had been coding since age 13, getting my first programming job two years later, and eventually, I studied computer science at Stanford—so, I’ve been engaging with computers for most of my life.
Learning about quantum computers and their capabilities was as magical as learning about computers at the age of six. It’s a fundamentally new paradigm for computing, and working on quantum computing seemed like the perfect chance to be part of a new era in computing.
Luckily, I met my now co-founder, Hrant, a physicist who has been studying quantum physics for a long time. We teamed up to found BlueQubit, making it easier for people to access and harness the power of quantum computers at the click of a button.
How Can People Access Quantum Computers Through Your Platform?
Quantum computers are like magical devices from the future that can run fundamentally different types of algorithms and computations than classical supercomputers. And they are becoming quite popular—many people have heard of them as a buzzword. Even Netflix features quantum computers in its shows, so they’re becoming mainstream.
Yet, using a quantum computer today is hard and requires expert knowledge in low-level programming, hardware engineering, or quantum physics. With BlueQubit, we are lowering the barrier to access quantum computers. We want quantum developers to get started at the click of a button and enable them to solve real-world problems in the near future.
When people talk about quantum advantage, quantum utility, or quantum supremacy, they are referring to the same thing: solving computational problems on a quantum computer that no classical computer could solve practically. That’s where we want to get.
One important part is benchmarking quantum computers and understanding their capabilities, not only for researchers and universities but also for companies and end customers. Defining a benchmark involves specifying a computational problem and an algorithm to solve it, running it on a quantum computer, and tracking metrics like speed, accuracy, or memory usage.
There are local metrics, such as a quantum computer’s two-qubit gate fidelity or coherence times T1 and T2. And there are high-level metrics like the quantum volume—how large of a quantum algorithm a quantum computer can run—and the accuracy of the results it’s producing. Our focus is on benchmarking high-level metrics; nevertheless, tracking both high-level and local metrics is important for eventually solving business problems.
Today, quantum computers are based on different kinds of platforms, with superconducting qubits and ion trap qubits being the most common ones, though others, like cold atoms or silicon spin qubits, are promising, too. Different platforms have different advantages and disadvantages, but generally, the number of qubits continues to increase, and, even more importantly, the quality of individual qubits and quantum gates is improving.
We expect the metrics to improve to the point where we will have a solid quantum advantage in a few years—you never know if it was a turning point unless you look back, but the fact that key technological milestones are being reached gives us confidence that we will get there.
What Applications of Quantum Computing Are You Excited About?
Quantum computers are particularly suited for doing computations to understand other quantum systems, so naturally, molecular dynamics and quantum chemistry applications make a lot of sense. That’s because, using normal computers, you need to make many approximations that work sometimes but not always, whereas with quantum computers, you can express those computations in a quantum-native way and run them directly.
For example, quantum computers will be able to simulate many different molecular bindings in minutes, which would take a classical computer years to compute. Obtaining such a quantum advantage is the holy grail of quantum computing.
Another interesting domain where quantum computers will have an outsized impact is in solving optimization problems. You don’t even need particularly many, but high-quality qubits. Even with just 100 high-quality qubits and about 10,000 two-qubit gates, you could achieve results no classical computer could match.
They will also be useful for Monte Carlo simulations, problems in graph theory, such as the maximum independent set problem, and quantum machine learning. People in academia often don’t like quantum machine learning—it’s not only two buzzwords instead of one, but you can’t prove things with pen and paper. Instead, you have to do many numerical experiments to figure out what works. But that’s the same with machine learning—there’s no rigorous mathematical theory of neural networks, and still, in praxis, they work.
Our platform offers the largest free quantum simulator, where developers can simulate running a quantum algorithm involving up to 34 qubits. They don’t have to worry about setting up and managing a quantum simulator themselves—they can simply submit the quantum algorithm they built, e.g., in Qiskit, and run them straightforwardly.
While it’s really hard to leverage AI to design quantum algorithms, and we might need to wait for some truly PhD-level artificial intelligence, quantum computers may already help run neural networks. We’re currently working on a simple, perceptron-style classification model that’s fully quantum and runs natively on a quantum computer. One of the major bottlenecks today is getting data in and out of a quantum computer, but fundamentally, neural networks are matrix multiplications, and quantum computers could help run them superfast.
How Did You Evaluate Your Startup Idea?
We’re seeing growing interest from the customer side, and no one doubts that quantum computing will be big. The main question is about timing and when we’ll get there. The technology advances continuously, unlocking new milestones, and more and more enterprises realize they need to start familiarizing themselves with quantum computing sooner rather than later.
Many companies missed out on the AI revolution and realized its potential only when ChatGPT made it obvious to everyone. Quantum computing is on a similar trajectory. Most of it is still being researched, but in the near future, quantum computers will show utility and unlock applications no classical supercomputer can tackle, leading to massive value creation that companies don’t want to miss out on.
What Advice Would You Give Fellow Deep Tech Founders?
Trust your gut—no matter how difficult you think it will be to found a startup, it will be more difficult and worth it only if you love what you’re doing and follow your passion. Don’t fall prey to the illusion that startups are a way to get rich quickly. I would work on quantum computing even if there wasn’t a prospect of making money. When you are founding a startup, you’re committing a significant portion of your life to a single mission—if you’re not confident you really want to do this, don’t do it.