Quantum Computing

What Is Quantum Computing?

Quantum computing is a model of computation that uses the rules of quantum mechanics to process information. Instead of bits that are either 0 or 1, it uses qubits, which can be placed in a superposition of both values and can be correlated with one another through entanglement.

The goal is not to speed up every program. It is to run particular calculations, such as simulating molecules and materials, at a cost that grows more slowly than any known classical method.

The idea dates to Richard Feynman, who argued in 1982 that simulating quantum systems efficiently might require a computer built from quantum parts. David Deutsch described a universal quantum computer in 1985.

How a Quantum Computer Works

A single qubit state can be written as

where \( \alpha \) and \( \beta \) are complex amplitudes. A register of \( n \) qubits is described by \( 2^n \) amplitudes, which is why classical simulation becomes costly as systems grow.

That does not mean the machine tries every answer and reports the right one. Measurement returns a single bit string, with probabilities set by the amplitudes. A quantum algorithm therefore has to be designed so that quantum interference cancels the amplitudes of wrong answers and reinforces the right ones.

Three ingredients do the work:

Superposition lets a register hold a weighted combination of many basis states.
Entanglement creates correlations between qubits that independent classical values cannot reproduce.
Interference shapes which measurement outcomes survive.

What Quantum Computers Can Do, and What They Cannot

Speedups depend on the structure of the problem, and the well-supported examples are few. Shor's algorithm (Peter Shor, 1994) factors integers in polynomial time, while known classical algorithms need superpolynomial time. No proof rules out a fast classical method, but none has been found.

Problem	Method	Type of speedup
Integer factoring	Shor, 1994	Superpolynomial over known classical algorithms
Unstructured search	Grover, 1996	Quadratic (about \( \sqrt{N} \) queries instead of \( N \))
Simulating quantum systems	Feynman, 1982; Lloyd, 1996	Expected to be large, depends on the system

Quantum computers are not general replacements for classical ones. Claims that they will accelerate all optimization or machine learning workloads are not established. In practice they are expected to work alongside classical hardware, which handles control, decoding and most of the workload.

Why Errors Define the Engineering Challenge

Qubits are fragile. Interaction with the environment destroys superposition (decoherence), and every gate is slightly imperfect. Hardware therefore needs constant quantum calibration, the tuning of control pulses so that intended operations are implemented accurately.

Today's devices are often called NISQ (noisy intermediate-scale quantum), a term John Preskill introduced in 2018. Their circuits are limited by noise rather than by the number of qubits.

Quantum error correction addresses this by encoding one logical qubit across many physical qubits and measuring parity checks to find and fix errors. The threshold theorem shows that if physical error rates are below a threshold, arbitrarily long computations become possible with a manageable overhead.

The overhead is large. Gidney and Ekerå (2019) estimated that factoring a 2048-bit RSA integer would take on the order of millions of noisy physical qubits running for hours, under stated assumptions. Better codes and hardware lower such estimates. This is why fault-tolerant quantum computing is the central milestone, and why QuEra has framed fault tolerance as its next milestone.

Where Neutral Atoms Fit

Quantum computers can be built from superconducting circuits, trapped ions, photons or atoms. QuEra builds with neutral atoms: identical atoms held in arrays by optical tweezers, interacting through highly excited Rydberg states. Because the tweezers can rearrange the array, the connectivity between atoms can be changed during a computation.

QuEra's systems use this in two modes:

Aquila is an analog system for Hamiltonian simulation, available on Amazon Braket since 2022.
Gemini is a gate-based system, on premises since 2025.
Libra is the fault-tolerant system, planned for Amazon Braket in 2028.

The Bloqade SDK supports work with these machines.

FAQ

Is a quantum computer faster than a classical computer?

Only for certain problems. Factoring and simulating quantum systems have strong theoretical support for large speedups, and unstructured search gains a quadratic one. For most everyday tasks, classical computers remain the right tool.

Can quantum computers break encryption today?

No. Running Shor's algorithm against modern keys needs large, fault-tolerant machines that do not yet exist. Organizations are already moving to post-quantum cryptography, and NIST published its first standards for it in 2024.

What is the difference between analog and gate-based quantum computing?

Analog machines evolve the qubits under a tunable Hamiltonian, which suits simulating physical systems. Gate-based machines apply a sequence of discrete operations and can in principle run any quantum algorithm. Both can be built from the same kind of atoms.

Will quantum computers replace classical computers?

No. They are expected to act as accelerators for specific workloads, with classical systems handling control, error decoding and most general computation.

Key Takeaways

Quantum computing uses superposition, entanglement and interference, and it helps only for problems with the right structure.
A quantum computer does not try all answers at once. Algorithms are built so interference favors correct outcomes.
Errors are the main engineering barrier, and error correction at scale is what enables long, reliable computations.
Neutral atoms held in optical tweezers are one hardware route, used in QuEra's analog and gate-based systems.