The founding story of BlueQubit is, by the standards of deep-tech startups, refreshingly unpretentious. In the spring of 2022, Hrant Gharibyan and Hayk Tepanyan were sitting on surfboards off Santa Monica when IBM put out one of its periodic announcements about progress on superconducting qubits. Most people who hear that news nod and go back to their day. These two paddled in and started a company. It is the kind of origin detail that sounds too neat to be true, except that both founders keep repeating it, which in the startup world is the closest thing to notarization.
What they built is worth understanding precisely, because the quantum industry has a habit of blurring the line between what exists and what is coming. BlueQubit does not build quantum computers. It builds the software layer that sits between a developer and whatever quantum resource they can get their hands on. The product is a Quantum-Software-as-a-Service platform - QSaaS, if you enjoy acronyms - and the pitch is almost aggressively practical: write your circuit in a framework you already know, such as Qiskit, Cirq or PennyLane, click run, and let BlueQubit figure out whether it lands on a GPU-based emulator or on a real quantum processor.
This is a more interesting business than it first appears. The dirty secret of quantum computing in 2026 is that the hardware, while genuinely improving, is still hard to use, expensive to access, and surrounded by tooling that assumes you have a physics PhD and a lot of patience. If you are a bank that wants to explore quantum Monte Carlo methods for pricing derivatives, the bottleneck is rarely a shortage of qubits somewhere in the world. It is that getting your quant team from an idea to a running experiment involves a stack of unfamiliar, unforgiving tools. BlueQubit's wager is that whoever makes that stack feel like an ordinary cloud API captures the customer long before the hardware is fully ready.
The emulator gambit
The clever part of the strategy is that BlueQubit does not need quantum hardware to be useful today. Its GPU-based emulators - built using NVIDIA technology - run quantum circuits on classical hardware, and the company claims they do so up to 100 times faster than commonly available alternatives. It offers a free 34-qubit simulator as an on-ramp, which is a generous number: each additional qubit roughly doubles the classical memory required, so the difference between a toy simulator and a serious one is a matter of exponents, not increments.
Free tools like this are often mistaken for generosity. They are really distribution. The graduate student who learns quantum programming on BlueQubit's free simulator is a plausible enterprise buyer in three years, and the habit of reaching for a particular platform is sticky in exactly the way software companies love. When the customer eventually needs a real quantum processor, BlueQubit provides managed access to those too - IBM's Heron and Quantinuum's H2 among them - through the same interface.
We decided to start the company while sitting on surfboards in Santa Monica, California, in the spring of 2022.Hrant Gharibyan, Co-founder & CEO
The founders' credentials do a lot of quiet work here. Gharibyan, the CEO, is an MIT graduate whose academic work included co-authoring an algorithm for so-called wormhole teleportation - a piece of theoretical physics that Google's Quantum AI team later implemented on a real processor. That is an unusual line on a resume: not just published research, but research that a company with one of the world's most advanced quantum programs decided to actually run. Tepanyan, the CTO, comes from the other side of the fence - a Stanford alumnus who worked on Google's infrastructure team, which is to say he knows how to build software that does not fall over when real users touch it. Quantum theory plus production engineering is a sensible pairing for a company whose entire premise is turning the former into the latter.
Money and the people who provided it
In December 2024, BlueQubit closed a $10 million seed round led by Nyca Partners, a firm known for fintech bets, which tells you something about where the company sees its first serious revenue. The syndicate is long and international: Restive, Chaac Ventures, Presto Tech Horizons, NKM Capital, BigStory, Untapped Ventures, Formula VC and Granatus Ventures among them. Granatus is notable - it is Armenia-focused, and BlueQubit is Armenian-founded, one of a growing cluster of technically ambitious companies with roots in the Armenian diaspora.
Tom Brown of Nyca framed the thesis in plainly commercial terms: the firm wanted a team positioned to help financial-services firms "hit the ground running once quantum is here." Note the tense. The investment is not a bet that quantum computing works perfectly today. It is a bet on who owns the developer relationship when it does - which is a much better bet, because you can win it early and the downside is a very good classical emulation business in the meantime.
BlueQubit builds first-in-class quantum software for current and upcoming quantum computers.BlueQubit
Advantage, with a bounty attached
The most striking recent development is scientific rather than financial. In October 2025, BlueQubit published a paper - "Heuristic Quantum Advantage with Peaked Circuits" - describing circuits deliberately engineered to be easy for a quantum computer and punishingly hard for a classical one. Running them on Quantinuum's H2 processor, with instances involving all-to-all connectivity and up to 2,000 two-qubit gates, the machine produced the target result in under two hours while classical simulation strategies choked. The technical claim is that the "peaked" structure pushes the point at which classical methods break down beyond what the team could reach even with heavy computational resources.
Quantum-advantage claims are a genre unto themselves, and healthy skepticism is warranted; "advantage" here is heuristic and problem-specific, not a general-purpose speedup. What is genuinely amusing is how BlueQubit chose to defend the claim. In February 2026 it launched a Quantum Advantage Challenge with a $20,000 Bitcoin prize for anyone who could classically beat its circuits. This is confidence expressed as a marketing budget. An open bounty turns skeptics into unpaid stress-testers and, if nobody collects, into unpaid endorsers. It is a good trick, and a slightly cheeky one.
The partner list is the other tell. An 18-person company should not, by the ordinary rules, be able to line up IBM, NVIDIA, Quantinuum, Honda, AWS, Stanford, Guidewire and QED-C. But being small and neutral is precisely the advantage. BlueQubit sits between hardware makers who want their machines used and enterprises who want a single door to knock on, and it is threatening to neither. Neutrality is a business model when the incumbents are each other's competitors.
The crowded doorway
BlueQubit is not alone in noticing that the software layer is where the money might be. The competitive set includes Classiq, which pitches high-level circuit synthesis; Q-CTRL, which focuses on error suppression; QC Ware and Multiverse Computing, which lean into applications; and Quantum Machines on the control side. And then there are the hardware vendors' own stacks - IBM's Qiskit Runtime, Amazon's Braket - which are simultaneously partners and potential competitors, the awkward reality of building on top of platforms that would also like to sell you the whole thing. BlueQubit's answer to this is to be the layer that speaks to all of them, which is a defensible position right up until one of the giants decides it wants that layer for itself.
What separates BlueQubit's positioning from the pack is less any single feature than a posture. The company publishes research, ships a free tier, and frames its mission as democratization - making quantum accessible to developers who are not quantum physicists. That word gets overused in technology, but here it maps to a concrete design choice: hide the ceremony. A developer should be able to reach for BlueQubit the way they reach for a cloud database, without first absorbing a semester of quantum information theory. Whether that abstraction holds as circuits get more demanding is an open question, but it is the right question to be organized around.
The unglamorous future
It is worth being honest about how quantum computing is likely to arrive, because it shapes whether BlueQubit's bet pays off. The popular imagination expects a single dramatic moment - a machine that cracks encryption or designs a room-temperature superconductor overnight. The more probable path is incremental and specific: a finance team pricing a class of derivatives slightly faster, a chemistry group modeling one more molecule than classical methods allow, a logistics operation shaving a percentage point off a routing problem. None of these make headlines. All of them make invoices. BlueQubit's product strategy - hybrid quantum-classical workflows, applications aimed at finance and materials science, a platform that meets customers on the hardware they already use - is built for that unglamorous, incremental version of the future, which happens to be the one that usually shows up.
The culture reflects the same seriousness. This is a small, research-driven team with an unusual density of pedigree - MIT, Stanford, Caltech, Google - and Armenian roots that connect it to a growing diaspora of technically ambitious founders. It behaves less like a company chasing a hype cycle and more like a lab that decided to also ship a product: peer-facing papers on one side, a working platform on the other. In a field that has burned through a great deal of investor and public goodwill on promises, doing the work in public is itself a differentiator.
None of this guarantees anything. Quantum computing has produced more disappointed timelines than shipped applications, and a software layer is only as valuable as the hardware and customers on either side of it. But BlueQubit has arranged its bets sensibly: a real product that works today, a research program that generates credibility, a fintech-flavored investor base that knows the first paying customers, and a free tier quietly recruiting the next generation of quantum developers. If quantum computing turns out to matter commercially, the company will have been standing in the doorway the whole time. And if it takes longer than the optimists hope, BlueQubit is still selling fast emulators to people who need them now - which is a perfectly good place to wait.