Oratomic's Remarkable $300 Million Series A
A new era of Quantum startups is starting in 2026.
Welcome Back,
While I wanted to cover IQM going public, an important Series A has arrived in 2026 out of the Quantum void, if you will.
I’ve always believed the Quantum industry would require fresh approaches to old challenges to make rapid progress. Oratomic (CalTech) a neutral atom quantum computing company that officially launched only this year, announced it has raised $300 million in its Series A. It has some unusual ideas on the future of Quantum.
Shor’s algorithm is possible with as few as 10,000 reconfigurable atomic qubits
The Oratomic origin story is rooted in a massive technical breakthrough that fundamentally shifted the timeline for fault-tolerant quantum computing (FTQC). What if 1 million Qubits is not required?
Even the investors willing to bet such moonshot feel different this time around - Backers include a round Co-led by ARCH Venture Partners, Spark Capital, and Khosla Ventures—with a heavy-hitting roster including Bezos Expeditions, Index, General Catalyst, and Lowercarbon Capital—this round stands out because of the company's aggressive, non-traditional strategy.
The Technical Epiphany
For years, the consensus in the quantum world was that running Shor’s algorithm (the math that can crack modern public-key encryption) would require millions of physical qubits to overcome hardware errors. Because of this massive error-correction overhead, utility-scale quantum computing always felt decades away.
Oratomic’s founding team flipped that script. Led by prominent quantum researchers and physicists—including Dolev Bluvstein, Manuel Endres, Harry Levine, and Caltech professor John Preskill, along with CTO Hsin-Yuan (Robert) Huang—the founders published a paradigm-shifting paper on arXiv.
The Core Breakthrough: By leveraging high-rate quantum error-correcting codes, flexible logical instruction sets, and dynamic circuit architectures, they proved that Shor’s algorithm could be executed at cryptographically relevant scales with as few as 10,000 to 26,000 reconfigurable neutral-atom qubits, rather than millions.
The Race to Fault Tolerance
So then Oratomic is building both the hardware and software for a large-scale fault-tolerant machine.
Their unique bet? By physically moving the atoms during the execution of an ultra-efficient error-correction scheme, they project they can achieve full fault-tolerance with just 10,000 to 20,000 physical qubits.
The background looks legit: Led by Chief Executive Dolev Bluvstein, the company launched in late March this year (March, 2026) to build utility-scale quantum technology using neutral-atom technology in collaboration with scientists at the California Institute of Technology.
Given the academics of Quantum are evolving fast now in the mid to late 2020s, you need entirely new teams to make new bets on the future. While many of the companies and startups I’ve covered over the last few years are finally going public, a new batch of startups are on their way.
On Light and Atoms
“At Oratomic, we are building the world’s first utility-scale quantum computers, enabled by a new regime of ultra-efficient error correction–using only light and atoms.”
Madelyn Cain (left) and Dolev Bluvstein (right) asked themselves: What is the smallest quantum computer one could imagine creating that could hack something like a Bitcoin wallet?
$300 Million is quite a sizeable Series A round and they are skipping steps, they are going to solve the key problem. The company plans to use the capital to expand quantum hardware fabrication, deepen research into fault-tolerant architectures, and grow its physics and hardware engineering teams. This is a star studded team we are talking about.
Quick Summary
By leveraging high-rate quantum error-correcting codes, efficient logical instruction sets, and a modular zoned circuit design, the researchers capitalize on the unique physical reconfigurability of neutral-atom arrays. Under their proposed architecture, a system utilizing 26,000 physical qubits could crack discrete logarithms on the P-256 elliptic curve in just a few days, while factoring RSA-2048 integers would take roughly one to two months. Overall, these findings significantly accelerate the projected timeline for utility-scale fault-tolerant quantum computing and underscore the urgent need to transition to post-quantum encryption standards.
Cain has the Key 🗝️
Oratomic is unusual in its speed to research - it emerged from stealth in March 2026 with groundbreaking research (in collaboration with Caltech) showing that cryptographically relevant quantum computing—e.g., running Shor’s algorithm on RSA-2048 or ECC-256—could be achievable with as few as ~10,000–26,000 physical reconfigurable atomic qubits, far below prior estimates of hundreds of thousands to millions.
Bluvstein and Cain had been working in the lab of the Harvard physicist Mikhail Lukin.
Rise of the Quantum Avengers.
The End Game
In an era of huge AI bets, it’s a time to aim for the holy grail of Quantum physics.
While they aren't selling intermediate SaaS or cloud access to early hardware, their ultimate end-game isn't just a standalone physics flex. Oratomic's stated goal for this fault-tolerant machine is to accelerate next-generation artificial intelligence systems and advanced scientific research pipelines, positioning them at the ultimate convergence point of advanced compute and AI architecture.
This is a Quantum Moonshot preparing for an eventful 2030s in Quantum architectures and computing.
Its stated objective is singular: it is not pursuing intermediate products or commercial systems along the way.
Let there be Light 🌈
Madelyn Cain is a prominent physicist and researcher specializing in quantum error correction (QEC) and quantum algorithms, specifically working with neutral-atom systems.
She completed her doctoral dissertation, “Quantum algorithms and quantum error correction with neutral atoms,” at Harvard University and has since been affiliated with institutions like the California Institute of Technology (Caltech) and the MIT-Harvard Center for Ultracold Atoms. She is also a founding team member at Oratomic, a quantum computing company focused on building fault-tolerant quantum computers using neutral atoms and light.
Low-Overhead Fault Tolerance: Her research introduces techniques to reduce the physical hardware and time overhead of quantum error correction by more than an order of magnitude, paving the way for more practical, utility-scale quantum processors.
Correlated Decoding: She has driven advances in “jointly decoding” logical qubits. By accounting for how errors deterministically propagate during transversal entangling gates, her work allows early fault-tolerant systems to drastically reduce the number of noisy error-checking steps needed.
Neutral-Atom Hardware Implementations: Her work heavily leverages dynamically reconfigurable neutral-atom arrays (using optical tweezers to move atoms around in real-time) to execute complex, error-corrected logical circuits and explore quantum adiabatic speedups in combinatorial optimization problems.
“We believe this will be one of our generation’s defining technological achievements. We also believe it can be realized with a small, focused team. This is possible because our recent advances have resulted in a substantially simpler approach to quantum error correction, based on reconfigurable atomic arrays trapped in focused laser beams.”
In the mid 2020s, the Neutral Atoms approach is in vogue.
As you might know, Co-founder Manuel Endres (Caltech) has already demonstrated neutral-atom arrays of ~6,000 qubits, providing a strong experimental foundation for scaling to the 10k+ range needed for utility. Neutral atoms offer long coherence times, good scalability (atoms can be trapped in large arrays), and reconfigurability advantages over fixed architectures like superconductors.
AI and Quantum
The problem involves frontier research and technology across advanced optical systems, electronics, atomic physics, mathematical algorithms, and error correction, all combined in a complex integrated fashion to realize this new type of computer. Their team are building artificial intelligence systems to automate their research and the design of the quantum computer.
In a Nutshell
Oratomic claims that they have found a smarter way to do this by using tightly focused laser beams, nicknamed “optical tweezers,” to pick up and move atoms around while they’re making calculations. Because they can catch and rearrange the atoms – the workers in this case – each worker can error correct more neighbors. Using this hardware, the company says it needs only 10,000 atoms to build qubits, down from a million.
So the triad in this zone of expertise is:
Atom Computing
QuEra Computing
Oratomic
These are immediately some of the frontier Quantum labs and startups to watch now in mid 2026, in my humble opinion. QuEra and Quantinuum are among the academic leaders. For an industry in need of a breakthrough, Oratomic is an interesting lean focused team.
The investors in this project feel different. In LinkedIn speak, Quantum computing has crossed from science project into the domain of the major funds. If that's not a validation of the scientific progress in quantum computing, I don't know what is.
We need more Quantum labs like this with bigger Series A rounds to justify their moonshots. Quantum deserves more funding, not just Generative AI. (which I do cover).
I’d be curious to know who the Angel investors are in this project.
ARCH Venture Partners
Spark Capital
Khosla Ventures
Bezos Expeditions
Index Ventures
General Catalyst
Lowercarbon Capital
Bain Capital
Formation
Nebular and more than a half-dozen others.
Nice to see CalTech on the map. But so many years to go.
Three months between a Seed round and a sizeable Series A tells you some pretty clear-eyed VCs think this has a good chance of becoming something important in the field.
The Evolution of Logical Qubits
As some of you might know, the 2030s are when logical qubits become way more useful. The company targets full fault tolerance from the start rather than noisy intermediate-scale quantum (NISQ) demos, aiming for utility-scale systems (thousands of logical qubits performing millions of operations) by the end of the decade. This could enable breakthroughs in molecular simulation, materials science, optimization, AI, and cryptography.
Don’t buy the 2030 hype, it’s going to take a lot longer than that. Companies are in the habit of pointing to the 2029 to 2030 sweet spot for various promises. But this stuff hard, you have to add at least five years to anything they say to appease onlookers.
The team also has experience combining AI with their Quantum experiments. Oratomic is also optimizing cryogenic chip packaging designed to support large-scale applications including machine learning and molecular chemistry simulations.
Additional Investors include:
David and Scott Aaronson
Les Kohn
Baiju Bhatt
Infleqtion
Genius Ventures
7i Capital and
Global Frontier Investments
The Calibre of the Team is Fairly High
CEO Dolev Bluvstein and team draw from Caltech, Harvard, Berkeley, Amazon, and Google expertise. Advisors/include John Preskill (a leading quantum information theorist). The launch paper involved top researchers, and the work has been highlighted by figures like Preskill.
“Oratomic’s founding team all previously believed that commercially useful quantum computing was far away,” Bluvstein said. “Our new research advances simultaneously changed all of our minds. We have assembled a team of top experts across neutral-atom quantum computing, error-correction theory, artificial intelligence, and optical engineering, and we are on a focused mission to build a utility-scale quantum computer.”
For a Series A, this is just what you want to see. Unbridled enthusiasm, unrealistic timelines and a decent size of funding.
All laughs aside, partnerships (e.g., with Monarch Quantum for photonics/systems integration) address control and manufacturing challenges, aiming for mass-manufacturable systems, potentially even room-temperature operation in some aspects.
“Pick a cooler life quest than building the world’s first quantum computer with your friends!” said a jubilant Bluvstein.
Obviously we still don’t know which Qubit approach is the best to reach scalability and fault-tolerance however instead of etching qubits into superconducting chips, the path Google and IBM have taken, this approach traps individual atoms with focused laser beams and uses them as the bits of the computer. It is cheaper to scale, and it is proving to be one of the most direct routes to a fault-tolerant machine.
If you don’t need Quantum computers with millions of Qubits, maybe the 2030s is the decade where this can finally happen.
The team proposes a new quantum error-correction architecture that is significantly more efficient than previous approaches. Say no more! Give them the funds, let’s get to work. Fault-tolerant quantum computing will change the world, it’s only a question of when and how combined with new architectures and computing paradigms of AI.
At a time of Quantum Summits, and meme-like IPOs of Quantum companies with little to no revenue, we need a more solid fast forward. CalTech is up to some interesting stuff. And now we have a Lab to cheer for with decent funding.
Inside Endres Lab
“"This is an exciting moment for neutral-atom quantum computing," says Manuel Endres, professor of physics at Caltech. "We can now see a pathway to large error-corrected quantum computers.” - Source.
Brownie points for having multiple female major contributors. (Yeah, I noticed!)
This will likely end up being the top U.S. Quantum moonshot startup of 2026. Not bad for a 4-month old startup.