Search the notebook
Follow a name, a question, an idea.
Search articles, notes, publication reviews, and news reports. Combine words with company, topic, and content filters.
Browse all articles, notes, publication reviews, and news reports
Control, algorithms and the work of scaling
The systems around the qubit come into focus: control, the cost of quantum algorithms, and the connections needed to scale. With selected research from IBM, NTT, and JPMorgan Chase.
The $775 million week
Six takeaways from quantum computing’s industrial buildout, 5–12 September 2026: CHIPS funding, foundries, software, control loops and the milestones still ahead.
Breakeven Is Where the Race Starts
A weekly-paper notebook entry on IonQ's breakeven demonstration of high-rate qLDPC codes: nine codes on forty barium ions, measurement without moving ions, what a correction cycle costs, and why the larger codes did not do better.
Not How Much Noise, but Which Kind
Two papers from Quantinuum and Pasqal show why noise amount sets reach, error type shapes the answer, and the dominant limitation changes with workload.
A Model Cannot Know What You Missed
A weekly-paper notebook entry on where noise models run out: Quantinuum's thermal-state experiment missing its own consistency check by six standard deviations, and Pasqal's cross-device model meeting a systematic offset it did not contain.
Nothing Left to Check Against: When Classical Verification Runs Out
A weekly-paper notebook entry on IBM Research and Algorithmiq's operator Loschmidt echo experiment: why benchmarking a quantum result against a classical simulation is circular, how an unverifiable estimate is made defensible, and why the method carrying real error bars runs only where a classical reference still exists.
The Answer the Classical Methods Could Not Settle
A weekly-paper notebook entry on the Qedma, IBM, RIKEN and BlueQubit driven-magnet experiment: a subharmonic rhythm that survives as the system grows, an inference the classically simulable sizes could not settle, and what the error mitigation carrying it actually costs.
Cheap to Believe, Expensive to Run: What It All Adds Up To
A weekly-paper notebook entry on IBM Research and the University of Chicago's doped Clifford sampling experiment: a fidelity certificate built into the circuit rather than inferred from a noise model, what it proves, what it costs, and what all three of IBM's advantage claims add up to.
Still the Referee: A Number Nobody Can Use
A weekly-paper notebook entry on IBM Research and the University of Chicago's LASSQD, a hybrid quantum-classical workflow for strongly correlated chemistry: what the quantum processor actually contributes, why connectivity matters more than qubit count, what the four validation layers show, and why the one calculation beyond exact classical fragment diagonalization cannot name a ground state.
What Is a Quantum Observable?
What a quantum observable is, and how it differs from a measurement outcome and from an expectation value. Hermitian operators and their eigenvalues, why a Pauli-Z measurement of the plus state averages to zero although no run returns zero, how bitstrings become eigenvalue samples, and why one energy estimate can take millions of circuit executions.
D-Wave and the Classical Counterattack
A weekly-paper notebook entry on the Science paper by Tindall and colleagues that reproduced an important part of D-Wave's quantum spin-glass benchmark classically using lattice-matched tensor networks and belief propagation: what it reproduced, what it left open, how D-Wave answered, and why quantum computing may not be a winner-takes-most market.
Errors as Telemetry: How Google Taught Willow to Tune Itself
A weekly-paper notebook entry on Google Quantum AI's reinforcement-learning control of quantum error correction on Willow: error-detection events repurposed as a training signal that re-tunes the machine as it drifts, what it delivered, what it costs, and the investor's read on Alphabet's full-stack position.
The Memory Has to Survive the Protection
A weekly-paper notebook entry on Quantinuum's first end-to-end error-corrected computation of a molecule's energy: quantum phase estimation on a color code, partial fault tolerance, and why memory noise, not the gates, was the limiting error in this circuit.
Quantum Phase Estimation: The Trick That Turns Physics Into a Number
What quantum phase estimation is: the algorithm that reads a fixed quantity, like a molecule's energy, off a quantum computer by hiding it in a phase and reading it back out. The textbook inverse-QFT method, the leaner single-ancilla information-theory version, and the first error-corrected demonstration, computing molecular hydrogen's energy on a trapped-ion machine.
A Gap, or a Crowded Room? Microsoft's Majorana Dispute
A weekly-paper notebook entry on the published Nature dispute over Microsoft's topological-qubit evidence: Henry Legg's critique, Microsoft's reply, and what a crowded gap means for the Majorana bet.
The Majorana Qubit: A Bit Hidden Where Noise Cannot Reach
What a Majorana qubit is: a qubit that stores its information non-locally across two Majorana zero modes, so local noise has a much harder time reading or corrupting it. How it is built and run by measurement, why quasiparticle poisoning is the enemy, why it remains contested, and why it is Microsoft's scaling bet.
The Random Telegraph Signal: Why a Good Qubit Won’t Stay Good
What a random telegraph signal is: noise that jumps at random between two fixed levels, the fingerprint of one microscopic two-level flaw switching back and forth. How many fluctuators add up to 1/f noise, how a single defect makes a qubit's lifetime drift, and why it unsettles every benchmark.
The Topological Gap Protocol: A Test Built to Not Fool Itself
What the Topological Gap Protocol is: Microsoft's pre-registered pass/fail test for deciding whether a device has truly become topological and hosts Majorana zero modes. How its three-terminal conductance test rejects look-alike Andreev signals, the gap values devices reached, and why it matters for a topological qubit.
Three Atoms, One State: The Promise and the Price of Networked Qubits
A weekly-paper notebook entry on the first fully distributed three-node GHZ state shared across remote atomic qubits (Goetting et al., 2026), what it cost in fidelity and rate, and what it means for IonQ's networking story.
Mermin’s Inequality: How to Catch a Computer Being Quantum
What Mermin's inequality is: a multi-particle test that scores measurements on entangled qubits and proves a system is genuinely quantum when it beats the classical limit. How the quantum gap grows with more qubits, the IBM five-qubit results, and why it is an honest benchmark for real hardware.
The GHZ State: All Together, or Not at All
What a Greenberger-Horne-Zeilinger (GHZ) state is: an all-or-nothing superposition of all-zeros and all-ones across many qubits. Its role in Bell's theorem without inequalities, its fragility, Heisenberg-limited sensing, and large GHZ states as benchmarks of global coherence.
The Quantum Circuit: A Recipe Written in Gates
What a quantum circuit is: horizontal wires are qubits, boxes are gates, read left to right in time and ending in measurement. How a universal gate set builds any circuit, why width, depth, and noise limit it, and why every quantum algorithm is in the end a circuit.
Measurement: The Price of Looking
What measurement does in quantum computing: it forces a qubit's blend of possibilities to collapse into one definite 0 or 1, set by the Born rule and lost to back-action; how dispersive readout and readout fidelity work; and why measurement can even drive a computation.
The Physical Qubit: The Fragile Raw Material
What a physical qubit is: an actual two-level quantum system built in hardware, what the DiVincenzo criteria require of it, how today's platforms compare on coherence and fidelity, and why no single physical qubit is yet good enough on its own.
Quantum Gates: A Solved Idea, an Unsolved Machine
What a quantum gate is, why every one has to be reversible, how a handful of gates can build any computation, and why gate fidelity rather than clever theory is the real wall to a useful quantum computer.
The Billion-Qubit Paradox: Why Your Quantum Computer Needs “Holes” to Think
Why a useful quantum computer needs thousands of physical qubits for every logical one, how the surface code computes by carving holes into a lattice and braiding them, and why factoring a 2,000-bit number would take roughly a billion physical qubits.
The Logical Qubit: Safety in Numbers
What a logical qubit is: one reliable unit of quantum information spread across many fragile physical qubits and protected by error correction, how the surface code builds it, what it costs in overhead, and why break-even is only the first milestone.
Entanglement: The Bond Einstein Bet Against
What entanglement is: particles sharing one joint state with correlations stronger than anything classical, how Bell's theorem turned Einstein's spooky-action objection into a test quantum mechanics won, why it is fragile, and why it is the engine of quantum computing.
Superposition: The Particle That Interferes With Itself
What superposition is: a quantum system genuinely in many states at once until measured, the single-electron double-slit and Schrodinger's-cat evidence for it, why decoherence makes it fragile, and why it is the engine of quantum computing.
Fault Tolerance: Computing Right Through the Noise
What fault tolerance means: running a long quantum computation correctly even though every part is noisy, why it is more than error correction, how the threshold theorem makes it possible, and why it is the dividing line to a useful machine.
The Qubit: Both Answers at Once, and Hard to Hold
What a qubit is: a unit of information that can be both 0 and 1 at once, how entanglement makes many qubits exponentially powerful, and why decoherence makes a good qubit so hard to keep.
Quantum Error Correction: Fixing What You Cannot Look At
Quantum error correction fixes mistakes you may not copy and may not even look at: how spreading one logical qubit across many physical ones plus syndrome measurements does it, why the threshold decides whether scaling helps, and why it is the bridge to a useful machine.
Qubit Connectivity: The Cost of Distance
What qubit connectivity is: which qubits on a chip can run a two-qubit gate directly, why limited connectivity forces costly SWAP gates that deepen and add noise to a circuit, the trapped-ion versus superconducting evidence, and why a bare qubit count can mislead.
Coherence Time: The Clock Every Qubit Races
What coherence time is: how long a qubit holds its quantum state before noise scrambles it, the T1 and T2 timescales, why decoherence causes it, how platforms range from microseconds to over an hour, and why it only matters next to gate speed.
Quantinuum: A Decade in the Lab, a Day on the Tape
A narrative-and-numbers company analysis of Quantinuum, told as a notebook.
Majorana 2: The Quantum Bet Microsoft Can Afford to Lose
A narrative-and-numbers company analysis of Microsoft's quantum bet, told as a notebook.
Qubit Count: The Scoreboard That Isn't
Why a quantum computer's qubit count is a poor score: it measures only width, ignoring gate fidelity, connectivity, and circuit depth. How quantum volume and volumetric benchmarks measure a machine honestly, and why quality, not quantity, is the real progress.
How Should Investors Read a Gate-Fidelity Claim?
A gate-fidelity number is not a complete score for a quantum computer. The questions to ask before trusting one: which gate, measured how (randomized benchmarking), on what system, and at what scale.
What Is Gate Fidelity?
What gate fidelity measures: how closely a physical quantum gate reproduces the ideal operation it is meant to perform, and why the step from 99.9% to 99.99% is a tenfold cut in error.
Xanadu: Building a Quantum Computer Out of Light, and Against the Clock
A narrative-and-numbers company analysis of Xanadu (XNDU), told as a notebook.
Show Me the Code: The Long Vindication of D-Wave
A narrative-and-numbers company analysis of D-Wave, told as a notebook.
The Stock That Went Shopping: The Quantum Computing Inc Story
A narrative-and-numbers company analysis of Quantum Computing Inc (QUBT), told as a notebook.
Rigetti: The $7 Million Company With an $8 Billion Story
A narrative-and-numbers company analysis of Rigetti Computing (RGTI), told as a notebook.
Qubits, and Why Quantum Computing Is a Winner-Takes-Most Bet on Physics
A physicist's tour of the four physics fields that today's qubits come from, and why the choice of physics is itself the bet investors are making.
IonQ: The Quantum Company That Decided to Become an Industry
A narrative-and-numbers company analysis of IonQ, told as a notebook.