This Weekend Bonus follows six stories about moving quantum computing from research toward usable systems, from circuit design and error modelling to hybrid infrastructure and industrial programmes. The achievements and plans are distinct: resource estimates, scheduled deployments and awards each tell us different things about progress.
Chapters
00:00 Today's headlines
01:07 QuiX Quantum's Dutch photonic mission
04:27 AI designed circuits reduced gate counts
07:24 IonQ announced executive changes
10:13 Rigetti Constellation digital twin
12:42 SEAFOOD hybrid infrastructure at Osaka
15:43 Jiuzhang 4.0 received leading prize
Quickly Quantum is an AI-voiced podcast, built and run by a real person. Nothing in this episode is financial advice.
Follow Quickly Quantum: subscribe on YouTube · Apple Podcasts · Spotify
More from Brian Lampert: Concrete Compute, the daily AI infrastructure briefing, and Space Stakes, the business of the new space race. Transcripts and every episode: quickly-quantum.kngoworld.chatgpt.site.
The daily quantum computing briefing you don't need a physics degree to follow. Every day, Quickly Quantum cuts through the hype to bring you the breakthroughs, funding rounds, policy moves, and research that actually matter, with plain-English analysis a smart non-physicist can follow. Hosted by Brian Lampert. AI-voiced, human-built, always skeptical of press releases. Nothing on this show is financial advice.
Also from Brian Lampert: Concrete Compute and Space Stakes.
A 100+ qubit photonic quantum computer is the target of a Dutch national mission, with nine partners working across the components needed to build it. The number sounds like a machine already taking shape, but this is a development programme, and its schedule stretches through December 2028. Who has to make the pieces work together before that target means anything? We’ll get into that, alongside the week’s other stories about AI-designed circuits, supercomputer links, error modelling and the people and programmes behind quantum’s industrial push. This is a Weekend Bonus: the other big quantum stories from the week, connected by a shift from lab results toward the capacity to build and deploy. There are two threads today: reducing the resources quantum computing needs, and building the industrial and computing infrastructure around it. Welcome back to Quickly Quantum, your daily brief on the quantum frontier. It's Saturday, November 14, 2026. Let's get into it.
QuiX Quantum’s Dutch programme puts the industrial supply chain at the centre of the 100+ qubit target, according to HPCwire. The company received funding for its Photonic Quantum Computing Mission under Phase II of the PhotonDelta National Growth Fund Integrated Photonics Programme. The mission runs from January 2026 through December 2028 and brings together QuiX Quantum and several Dutch industrial and academic partners. Together they will develop and integrate twelve deliverables covering the quantum photonic stack, from low-loss chips and photon sources to detection, control electronics, packaging and manufacturing.
That’s the first thread in miniature: a quantum computer depends on a chain of components, and a target for the finished system also depends on those pieces working together. Photonic quantum computing uses particles of light to carry quantum information. Here, the programme’s work ranges from making light travel through chips with less loss to generating and detecting photons, then controlling and packaging the system. The collaborators include imec on silicon nitride waveguides and high-speed modulators; PHIX and MicroAlign on fibre-to-chip coupling; the University of Twente on a quasi-deterministic single-photon source; and Eindhoven University of Technology on photon distillation and resource-state generation. Leiden University is working on on-chip photon-number-resolving detectors, while Eindhoven University of Technology is also working on integrated SPAD arrays.
The control work comes from Bruco-IC, which is targeting latencies below 10 ns for flip-chip control ASICs, compared with 150 ns for today’s feed-forward control. Latency is the time taken for a control signal to respond. In this system, faster responses matter because the machine’s control has to keep up with its quantum operations. The consortium also targets inverse design and post-fabrication tuning to improve chip tolerances and yield, meaning the proportion of manufactured chips that meet requirements.
HPCwire says QuiX leads the mission, integrates partner results at system level and tracks progress against a shared KPI and technology-readiness framework. The mission concludes with a manufacturing strategy for 100+ qubit systems. That is a plan for industrialisation, not an account of a finished 100+ qubit machine.
There’s also an earlier step behind the target: QuiX delivered its Universal Photonic Quantum Computing hardware platform, developed for the German Aerospace Center’s Quantum Computing Initiative, earlier in 2026. The company is advancing a next-generation Fault Tolerant architecture, with development and demonstration of loss-tolerant logical-qubit capabilities planned throughout 2028–2030. Logical qubits are the error-protected units a fault-tolerant machine would use; the company says these capabilities would lay the foundation for commercially deployable fault-tolerant photonic quantum computers beyond 2030.
I think the useful signal here is the breadth of the work, because the programme is trying to coordinate components that have to function as one system. The counterweight is the distance between funded development and demonstrated fault-tolerant operation. I’d be more persuaded by progress on the logical-qubit capabilities than by the size of the target alone. For now, the concrete outcome is a mission with defined deliverables, named technical work and an industrial goal. The 100+ qubit figure is a destination, and December 2028 is when this mission concludes.
Meanwhile, the second thread begins inside the circuits themselves: researchers from QuEra Computing, Harvard University and Anthropic used Claude to design circuits that reduced gate counts and qubit requirements across seven of eight computational building blocks, The Quantum Insider reports. The work is described in the preprint “Discovering state-of-the-art quantum circuits with large language models.”
A quantum circuit is a sequence of operations that carries out a computation. The team paired Claude with an independent automated evaluator, which checked whether each proposed circuit was correct and scored its resource use. Claude used that feedback to explore and improve its designs. The evaluator measured Toffoli-gate count and qubit count. A Toffoli gate is a particular kind of operation in a quantum circuit; the paper says these gates can account for a substantial share of the overhead in fault-tolerant computations. Logical qubits, meanwhile, occupy valuable hardware capacity.
The results varied by task. For a Reed-Solomon decoder, which the article describes as the most expensive step in decoded quantum interferometry, the AI-designed circuit used 12x fewer Toffoli gates, from 5.1 million to about 440,000, while also using fewer qubits. For a Coulomb potential calculation, a way of describing how charged particles interact when simulating molecules, the combined resource metric fell by 9.9x; Toffoli gates went from 6,355 to 1,140 and qubits from 252 to 142. The largest reported reduction in that combined metric was 39x for a fixed-point Gaussian function, with Toffoli gates down from about 24,000 to 1,583 and qubits from 362 to 139.
The 39x figure needs its precise meaning: the combined metric multiplies the Toffoli-gate count by the peak number of logical qubits, so it does not mean a 39x speedup. Runtime and physical hardware requirements depend on the system’s architecture. These results are resource estimates for individual building blocks rather than runs on hardware, and their effect on complete applications will depend on how those components are used.
The eighth building block is also interesting: the search found a new approach offering additional choices between qubit requirements and gate counts. The article says the team also developed a new reversible approach to modular inversion, an arithmetic operation central to elliptic-curve calculations. This is a trade-off, not a single design that wins on every measure.
I think the strongest part is the method’s use of an evaluator to check correctness while the AI searches for lower resource use. That gives the result more substance than a model proposing circuits without a way to assess them. But the evidence remains at the level of circuit estimates, and the complete-application effect depends on how these blocks are combined. The result strengthens a practical idea: better algorithms can make limited hardware go further. It doesn’t yet tell us how much a full quantum application’s runtime or physical requirements will change.
A different kind of capacity question arrived at IonQ, where the company announced executive changes that it says are intended to position it for continued growth and advance its path to profitability. The company’s announcement names Gary Merrill, currently Chief Financial Officer of Commvault Systems, Inc., as IonQ’s new Chief Financial Officer, effective early December. Thomas A. Austin, formerly Chief Accounting Officer at Juniper Networks, an HPE company, becomes the newly created Chief Accounting Officer, effective immediately. Inder M. Singh is stepping down from his roles at the company, effective immediately. The announcement is from IonQ; this show has not independently confirmed the appointments.
IonQ’s chairman and CEO, Niccolo de Masi, said the new finance leaders bring experience that will matter as the company scales and works toward profitability. He said Merrill’s public-company CFO experience and cybersecurity expertise would be particularly valuable. Merrill’s background spans finance, operations and go-to-market work, according to IonQ. The company says he returned to the CFO role at Commvault in April 2026 after serving as its Chief Commercial Officer from 2024 to 2026. Before that, he was CFO from 2022 to 2024.
For a listener new to IonQ, it builds trapped-ion quantum computers: electrically charged atoms held in place are used to store and process quantum information. This announcement is about the company’s executive team and financial leadership, not a change to that computing approach.
Merrill said: “Throughout my career as an operator, commercial leader and CFO, I have seen firsthand the importance of pairing ambitious growth plans with disciplined financial management.” He also called IonQ’s balance sheet “a distinct advantage in the quantum industry.” Those are Merrill’s claims about the company. I take the first sentence as a clear statement of the job he sees ahead: balancing growth with financial discipline. The second is his assessment, and the announcement offers no figures to evaluate it here.
The one public reaction in the supplied discussion is a short headline from @QuantumStateX on X: “IonQ says new CFO brings cybersecurity expertise to quantum firm”. That captures the detail IonQ itself emphasizes about Merrill, but it doesn’t offer an argument on the appointment or the company’s financial prospects.
I think a CFO appointment is a meaningful business signal because it puts financial management in the foreground while IonQ says it is pursuing growth and profitability. Still, personnel experience can’t establish that those goals will be met. Merrill’s background at a public company and his work in cybersecurity are relevant credentials, while the company’s stated ambitions remain its own. The change gives IonQ a new finance leadership structure; the company’s progress toward profitability will have to be judged by outcomes rather than the titles announced today.
Another piece of the capacity puzzle is modelling: Rigetti Computing integrated Quantum Elements’ Constellation simulation environment into its quantum error-correction workflow, Quantum Computing Report says. Error correction is the process of detecting and managing errors so a quantum computer can perform reliable calculations. After a three-year technical collaboration, Rigetti validated Constellation’s ability to predict single- and two-qubit gate fidelity across its 36-qubit and 108-qubit quantum processing units.
Gate fidelity describes how closely a quantum operation matches the operation intended. The simulation environment, which Quantum Computing Report calls a digital twin, runs physical simulations of target processors on classical computers. It models gate-level fidelity and error channels, including decoherence and cross-talk. Decoherence is the loss of delicate quantum behaviour; cross-talk is unwanted influence between parts of a device. The platform generates calibrated noise-model parameters, which feed estimates of logical error rates at scale.
Why should you care about a model rather than a new processor? The report says the workflow lets engineering teams evaluate pulse schedules and system dynamics without consuming physical quantum-processing-unit time. It also describes hardware-aware gate scheduling and optimisation before physical runs. In principle, that lets engineers use simulation to investigate processor behaviour before spending time on the device. Quantum Computing Report says the work gives Rigetti’s hardware and error-correction teams a unified benchmark for diagnosing performance-limiting mechanisms as qubit counts grow, citing CEO Subodh Kulkarni.
The specifics matter: the validated prediction concerns single- and two-qubit gate fidelity on the 36-qubit and 108-qubit processors. The published description also says the modelling reaches beyond small few-qubit subsystems, capturing decoherence and cross-talk behaviour. Those are useful details about what the workflow is intended to model; they aren’t a reported logical-qubit demonstration.
I think this is a practical engineering step, because error correction depends on understanding the physical errors that the system has to handle. A model that helps engineers explore device behaviour before a physical run can make that work more targeted. But prediction of gate fidelity and an estimate of logical error rates are different from showing that a logical qubit operates successfully on hardware. The announcement moves the modelling into Rigetti’s workflow; its value will depend on how well the predictions help teams diagnose and improve the actual processors.
On the deployment side, the University of Osaka has developed SEAFOOD, a hybrid infrastructure linking its OCTOPUS supercomputer with domestically built superconducting quantum processors across separate campuses, according to Quantum Computing Report. Hybrid computing here means dividing a job between classical computers and quantum processors, with each handling the parts suited to it. SEAFOOD is scheduled to enter formal operation in December 2026. Quantum Computing Report describes it as the first integration of a Japanese-developed quantum computer and an on-campus supercomputer at a single academic institution in Japan.
The system has three layers. OCTOPUS supplies classical high-performance computing, including the matrix diagonalization used in the reported workflow. The quantum layer includes superconducting processors at the QIQB site on Toyonaka campus, including Urchin, with chips supplied by RIKEN Center for Quantum Computing. The software layer uses OQTOPUS, an open-source quantum cloud stack co-developed with Fujitsu, SEC and TISI, to route circuit jobs and measurement results between the high-performance computing nodes and quantum processors.
There’s a concrete demonstration behind the architecture. In a Quantum-Selected Configuration Interaction demonstration, the quantum processor sampled dominant electron configurations, while OCTOPUS’s high-memory compute nodes constructed and diagonalized the resulting subspace Hamiltonian. A Hamiltonian is a mathematical description of a system’s energy; diagonalizing it is a way to calculate energy states. The example used a four-hydrogen molecule and 1,000 shots on Urchin, producing a ground-state energy of −2.1147334880 Hartree, matching classical CASCI reference values.
That split is the point: the quantum processor handles the sampling, and the classical supercomputer handles the construction and diagonalization work. The report connects strongly correlated electronic structures to drug discovery and materials engineering, while also noting the constraints on near-term quantum processors from qubit count and noise. This demonstration shows a hybrid workflow for a specific calculation; it doesn’t establish that such a setup has already made a drug-discovery or materials breakthrough.
The plan also sets a route for researchers to try the infrastructure: in fiscal year 2027, up to 10 research teams at The University of Osaka will receive trial access through the D3 Center’s proposal-based program to refine hybrid job scheduling, ahead of broader external access via Japan’s nationwide JHPC-quantum initiative.
I think this is a useful deployment story because it’s about how quantum hardware can be connected to computing systems researchers already use. The reported molecule calculation gives the architecture a tangible example, and the planned trial access brings researchers into the scheduling work. What this supports is a path for testing hybrid workflows, with broader external access still ahead. For you, the important detail is that the quantum processor is one part of a larger system, and the job is to make the handoff between those parts work.
For a lighter turn toward recognition and national strategy, Jiuzhang 4.0, a programmable photonic quantum computer, received the leading prize at the 2026 World Internet Conference in Wuzhen, Zhejiang Province, Quantum Zeitgeist reports. The awards recognized projects across basic research, key technologies and engineering development. The conference received over 430 submissions from 43 countries and regions, and 20 projects received the WIC Leading Science and Technology Awards.
The other winning projects included advances in optical transmission for space-based quantum information and intelligent vehicle agent technology. Quantum Zeitgeist says the awards aim to recognize technological advances worldwide and promote international exchanges and cooperation in internet technologies. It also reports that ten winning projects focused on key technologies, while eight demonstrated advances in engineering and development.
The award puts a photonic quantum computer in a wider set of technology achievements, but the recognition itself doesn’t tell us what technical milestone Jiuzhang 4.0 demonstrated. That distinction matters: an award can mark attention and institutional recognition without serving as a measurement of a machine’s performance. The article identifies the prize and its competitive field, but gives no technical performance result to weigh here.
I read this as a sign of the visibility China’s photonic quantum work has in a broad technology forum, rather than evidence on its own about computing capability. The other projects named alongside Jiuzhang 4.0 span very different areas, so the award gives a sense of what the conference chose to recognize, not a direct comparison between quantum systems. The useful next development would be a reported technical result tied to the machine itself; today’s news is the award.
If you’re enjoying these Saturday deep dives into the work behind quantum computing, follow Quickly Quantum wherever you listen. This has been Quickly Quantum, an AI-voiced podcast, created and built by a real human using today's cutting-edge technology. Nothing you heard on this show is financial advice. I'm Brian Lampert, and I'll catch you all tomorrow — take care!
Watching on YouTube? Hit subscribe, and tomorrow's briefing will be waiting for you. I also host Concrete Compute: a daily briefing on the AI buildout. The datacenters, the megawatts, and who actually pays for them. Find it wherever you get your podcasts.