Haiqu: Shaping the Future of Software to Boost Quantum Computers’ Performance
How can we make do with the current quantum computing hardware? It’s the multi-billion dollar question the industry is facing right now that, after decades of research, quantum computers still haven’t become commercially viable.
While some quantum startups keep pushing for the dream of fault-tolerant quantum computing, others take a more pragmatic approach to see what current noisy-intermediate scale quantum (NISQ) computers could be useful for – if not completing calculations orders of magnitudes faster, maybe they can do so more resource efficiently.
Haiqu was founded by Mykola Maksymenko and Richard Givhan in the fall of 2022 to build a software layer to enhance existing quantum hardware and make it perform several orders of magnitude better. It raised a $4M seed round last summer, led by MaC Venture Capital, with participation from Toyota Ventures, SOMA Capital, u.ventures, SID Venture Partners, and Roosh Ventures.
Learn more about the future of software to boost quantum computers’ performance from our interview with the co-founder Mykola Maksymenko:
Why Did You Start Haiqu?
Throughout my career, I have been working as a physicist in fundamental research and AI research, eventually leading the RnD department of a larger IT company. I had a pretty good and comfortable position in the industry, but at some point, I realized that founding a startup could give me yet another very different experience and take me to the next level. Once you reach such a point in life, it’s clear that you should give it a go.
I took a sabbatical from my previous role and joined the Quantum Stream of Creative Destruction Lab (CDL). It began with an intensive four-week quantum boot camp, which was a great learning opportunity, allowing me to explore a number of startup ideas in quantum computing and how those could be commercialized. Through the boot camp, I met my co-founder Richard, and we decided to tackle the elephant in the world of quantum computing: How to make NISQ quantum processors perform a lot better so they actually become commercially viable.
We finished the boot camp with an idea and a two-person team and got into the main part of the CDL program – five goal-setting sessions over eight months to sharpen objectives, prioritize time and resources, raise funding, and engage with researchers.
It helped us find prospective customers, get their feedback, and build a prototype. After each session, we got a set of objectives that we had to achieve by the next one. Through this process, we got strong interest from the market and understood how our technology could scale. I never returned from the sabbatical; instead, we started fundraising and eventually founded Haiqu.
How Can Software Boost Quantum Computers?
We’re developing a software layer to enhance existing quantum computers and make them perform several orders of magnitude better. In practice, this means to dramatically increase their quantum volume, allowing you to run much more complex quantum algorithms.
The quantum volume is a comprehensive metric that reflects the overall performance of a quantum computer. It measures the largest random quantum circuit of equal width (number of qubits) and depth (number of operations) that a quantum computer can successfully implement. Such circuits are some of the hardest to execute on modern quantum devices.
Simply put, when a quantum algorithm “fits” within the quantum volume dimensions, a quantum computer will likely execute it efficiently and with a high degree of fidelity. However, if it exceeds the quantum volume’s average number of operations or qubits, errors tend to accumulate rapidly, leading to unreliable results. Everyone is pretty hyped about quantum hardware startups pushing the qubit numbers of their machines, but more importantly, their quantum volume needs to increase too.
In the current quantum computing stack, you have the quantum processors that power everything at the bottom and the applications that end-users will be using at the top. In between sits middleware like quantum compilers, noise mitigation techniques, and software that optimizes the applications to run more optimally. And that’s exactly where we come in: our software enables a different way of running a quantum algorithm while still allowing you to use your favorite quantum compiler and noise mitigation techniques.
We’re actively exploring and advancing other frontiers of the middleware layer too. For example, we’re looking into the ideal quantum algorithm compilation, advancing noise mitigation techniques, and facilitating distributed quantum computing – imagine a network of smaller quantum computers, each connected to form a powerful, cohesive grid.
The big remaining question is: What can we do with quantum computers still having just a few hundred qubits but a much greater quantum volume?
From an academic perspective, they may assist fundamental research in theoretical condensed matter physics, e.g., to understand quantum phenomena such as many-body localization or non-equilibrium quantum systems. These are classically hard or even impossible to simulate, and near-term quantum computers may provide a path forward.
More practically, they can simulate quantum processes in quantum chemistry and improve the accuracy of energy structure calculations. This is relevant, e.g., for catalysts, as their reactivity sensitively depends on their energy levels. A better understanding of their reactivity will help to devise new synthesis routes for materials.
How Did You Evaluate Your Startup Idea?
Despite the best efforts of hardware providers, quantum computers today deliver no commercially relevant advantage over classical supercomputers, so it was clear to us that they’re looking for a solution that may improve their performance. And on the applications side, people struggle with their applications being too big and complex for current quantum hardware, i.e., they don’t have enough qubits and quantum volume to execute a quantum algorithm before it decoheres.
We talked to many application developers and found that Haiqu could be the unlock to bring both sides together: make their quantum algorithms run for long enough even on near-term quantum hardware and help them get the most utility from existing NISQ processors.
Now that we have our system up and running, we are preparing a white paper with benchmarks to demonstrate the value of our technology to selected customers. Therein, we considered several examples of different quantum processors and how our software will improve their capabilities. This includes running several standardized benchmarks and testing with application providers to obtain benchmarks for their very applications.
What Advice Would You Give Fellow Deep Tech Founders?
It is important to get early feedback, mentorship support, and a strong network when you are just about to launch a new venture. I recommend joining a good accelerator, incubator, or venture lab that will provide you with those in your particular domain. Participating in the Quantum stream of Creative Destruction Lab has helped us a lot and is a great first step if you’re coming from academia or are generally a first-time founder. It’s worth it not only for the experience, knowledge, and network but also because it will put you outside of your comfort zone, you’ll meet potential co-founders, and iterate on ideas jointly.
Work on the Future of Computing at Haiqu
As an interdisciplinary team, we’re on a mission to breathe life into NISQ (Noisy Intermediate-Scale Quantum) technology. Explore our job page for opportunities to join us in shaping the future of quantum computing*
Technical Product Manager*
Join Haiqu and be the driving force behind our innovative quantum middleware. Your chance to collaborate with top-tier engineers and researchers, understand market trends and guide our quantum computing products to success.
Research Scientist*
Dive into cutting-edge research, pioneer new quantum algorithms, and showcase your work on global stages. Join our team of Stanford and Weizmann Institute alumni and lead the charge in quantum algorithm development.
*Sponsored links – we greatly appreciate the support from Haiqu
Learn more about sponsoring the Future of Computing blog here