Learn
I want to understand quantum computing.
Best for · Beginners · students · developers
Start with one Bell-state circuit on a local simulator. Predict the result, run it, and explain what the measurements mean.
Quantum, from the beginning
Start with one small, testable workflow.
You do not need a quantum computer, provider account, or physics degree for the first step. Learn the basic model, run one circuit locally, and keep the result understandable.
- First
- Learn only the concepts needed to read one small circuit.
- Then
- Run locally and compare the result with your prediction.
- Later
- Use real hardware only when it answers a hardware-specific question.
Choose by intent
What are you here to do?
Pick one path. You can ignore the rest of the page until that path becomes relevant.
Evaluate
I need to evaluate a use case.
Best for · Research · enterprise teams
Define the question and classical baseline first. Move to real hardware only when physical-device behaviour can change the decision.
Secure
I need to prepare for quantum-safe security.
Best for · Security · infrastructure leaders
Treat post-quantum migration as a separate classical engineering programme: inventory, prioritise, pilot, and build crypto agility.
See the more specific learner, educator, developer, research, and enterprise routes
Complete beginner
Understand one circuit before collecting tool names
A first circuit you can predict and explain
Student or educator
Turn foundations into a four-week learning route
An assessable lab record and portfolio artifact
Software developer
Build an inspectable Qiskit workflow
Readable source with a simulator baseline
Research team
Make the experiment repeatable before making a claim
A reproducible experiment record
Enterprise team
Frame a bounded pilot with a classical baseline
A pilot that can support an investment decision
Security leader
Treat post-quantum migration as a separate programme
A risk-prioritized migration starting point
Your first 30 minutes
Predict. Run locally. Explain what happened.
IBM’s quickstart reviewed on 11 August 2026 uses Qiskit 2.5-compatible packages and a local StatevectorSampler. The first workflow needs no provider login or API key.
The circuit
q0 |0⟩ ── H ── ● ── M q1 |0⟩ ─────── X ── M
Expect mostly 00 and 11. A successful Bell run demonstrates the workflow; it does not demonstrate useful quantum advantage.
Predict
Write down the expected result before touching the run button: a Bell circuit should produce strongly correlated 00 and 11 measurements.
Keep · one-sentence prediction
Build and run locally
Create H plus CX, add two measurements, inspect the Qiskit source, and use a local StatevectorSampler before introducing a provider account.
Keep · circuit, source, and raw counts
Explain and save
Describe the distribution and sampling variation. Record the environment, settings, result, limitation, and one deliberate revision.
Keep · interpretation and evidence packet
Open the complete six-step workflow
- 01
Frame and predict
Write the question first: after a Bell-state circuit, you expect strongly correlated 00 and 11 measurements rather than every bit string.
Retain · One-sentence prediction
- 02
Build H + CX
Prepare two qubits, apply a Hadamard gate to the first, use it as the control for a CX gate on the second, then measure both.
Retain · Two-qubit circuit
- 03
Inspect the source
Find the circuit, registers, H, CX, measurement, shot count, execution route, and result object in the generated or written Qiskit source.
Retain · Reviewed source
- 04
Run locally
Use a local StatevectorSampler before creating a provider dependency. Repeat with different shot counts and keep the raw counts.
Retain · Simulator result record
- 05
Explain the distribution
Describe the expected correlation, sampling variation, and why a successful Bell run is not evidence of useful quantum advantage.
Retain · Short interpretation
- 06
Preserve and revise
Save the question, workflow, source, environment, route, settings, result, interpretation, limitations, and one deliberate change.
Retain · Reviewable evidence packet
Continue with a four-week learning plan
Week 1
Build the mental model
Bits and qubits, state, gates, circuits, measurement, probability, and the limits of common analogies.
Checkpoint · You can predict and explain a small circuit without code.
Week 2
Make circuits executable
Bell states, interference, shot-based results, local simulation, Qiskit structure, and OpenQASM inspection.
Checkpoint · You can connect a circuit diagram, source, and result record.
Week 3
Learn algorithm patterns
Oracle-based search, variational workflows, classical optimization loops, baselines, noise, and repeated trials.
Checkpoint · You can describe an algorithm without claiming automatic speedup.
Week 4
Produce a reviewable project
Problem framing, simulator validation, optional QPU qualification, run metadata, interpretation, limitations, and revision.
Checkpoint · You have a portfolio artifact with method, evidence, and limits.
Open the minimum vocabulary
- Qubit
- A quantum information unit described by a state. Measurement produces a classical result; it is not a classical bit that reveals both answers.
- Quantum state
- The mathematical description used to predict measurement probabilities. It changes through operations and cannot be read completely from one measurement.
- Circuit
- An ordered representation of quantum operations and measurements. Real execution also needs a target, compilation choices, and run settings.
- Shot
- One repetition of a measured circuit. Repeated shots estimate an outcome distribution but do not remove hardware or modelling error.
- Simulator
- Classical software that models quantum behavior. It is the first route for logic checks, though large simulations can become expensive.
- QPU
- Physical quantum hardware affected by supported instructions, topology, calibration, noise, queue, and access terms.
- Transpilation
- Rewriting a circuit to match a target's instruction set, connectivity, and optimization choices.
- Primitive
- A defined execution interface. SamplerV2 returns samples from measured circuits; EstimatorV2 estimates expectation values for circuit-observable inputs.
Real hardware comes later
Stay on the simulator until hardware changes the question.
A QPU is not the default run button. Promote a workflow only when topology, supported instructions, calibration, noise, queue, or another physical constraint matters to the result.
Default route · local simulator
Use it to learn and debug.
- Check the prediction, circuit logic, parameters, and result parsing.
- Keep a stable baseline without account, queue, or provider state.
- Revise freely while the workflow is still changing.
Promoted route · QPU
Use it for a physical question.
- Select a target and inspect the ISA-compatible circuit and layout.
- Record access, queue, usage, calibration context, and job identity.
- Define the evidence need and stop condition before spending more runs.
Compare the four simulator-to-QPU decision questions
01
Does the local workflow pass?
Simulator · The prediction, circuit, source, parameters, and result parsing work in a controlled local route.
QPU · Do not submit yet if logic, measurement, or result handling is still changing.
02
Is hardware behavior part of the question?
Simulator · Use local reference primitives for ideal logic checks; use a noise-aware simulator when a modelled device comparison is useful.
QPU · A QPU is justified when topology, ISA, calibration, noise, queue, or physical-device behavior can change the decision.
03
Is the target fit understood?
Simulator · Record circuit width, depth, observables, shots or precision, and the expected output contract.
QPU · Select a target, transpile to its ISA, inspect layout and depth, and confirm the primitive and execution mode.
04
Are access and stop conditions explicit?
Simulator · Keep the reproducible baseline and expected result before introducing provider state.
QPU · Verify account, instance, plan, region, queue, usage budget, data boundary, job retention, and the condition that stops more runs.
Understand Qiskit, IBM Quantum, and QFlow Studio as separate roles
Qiskit
Open-source SDK
Build circuits and operators, use local reference primitives, transpile for a target, and work with execution results.
Boundary · Qiskit does not itself grant an IBM Quantum account, instance, plan, allocation, or QFlow product capability.
IBM Quantum
Provider platform
Controls its platform, accounts, instances, plans, Runtime services, available QPUs, access terms, and hardware documentation.
Boundary · Availability and entitlement are provider-controlled and must be checked live before a lab or paid run.
QFlow Studio
Neura Parse product
Connects visual workflow intent, generated source views, local simulator routes, run history, Academy labs, and reviewer-safe evidence.
Boundary · It does not replace Qiskit or IBM Quantum. Provider routes remain status-labelled; specialized routes and qOS remain planned or in development.
Neura Parse is an IBM Partner Plus member. This does not imply IBM endorsement, IBM Quantum Network membership, certified QFlow integration, bundled or privileged QPU access, or research validation.
Choose a next surface
Product, research, and delivery answer different questions.
Pick the surface that matches the work. A product feature, research result, and guided programme do not inherit one another’s maturity or evidence.
Product
QFlow Studio
Commercial visual quantum workflow and learning product connecting canvas, generated Qiskit, Cirq and OpenQASM views, local simulation, run history, Academy, and evidence.
Public research
Guided delivery
Quantum Systems Engineering
Education programmes, repeatable research workflows, and bounded enterprise pilots with baselines, resource estimates, evidence gates, and handover records.
Inspect route descriptions, evidence boundaries, and sector programmes
Learn and build
QFlow Studio
Commercial visual quantum workflow and learning product connecting canvas, generated Qiskit, Cirq and OpenQASM views, local simulation, run history, Academy, and evidence.
Boundary · Current, beta, validation-only, preflight, credential-test, and planned routes remain explicitly separate.
Decision research
QANTIS
Community Edition research on quantum belief updates and data association, including two public 2026 arXiv papers reporting IBM Heron experiments.
Boundary · The latest study treats the QPU as a calibrated belief-update service inside a classical planning loop; it is not a total-runtime or quantum-advantage claim.
Research substrate
qmesh
Typed quantum IR, backend abstraction, signed manifests, hybrid scheduling, and fault-tolerant promotion research.
Boundary · qmesh composes with Qiskit, Cirq, and PennyLane; it is not positioned as their replacement.
Quantum–AI research
QMANN
Quantum memory-augmented neural-network research with theoretical, simulation, and hardware modes plus a cost check before hardware execution.
Boundary · Mode and cost boundaries must remain visible; a simulated or hardware run is not an automatic performance claim.
Applied delivery
Quantum Systems Engineering
Education programmes, repeatable research workflows, and bounded enterprise pilots with baselines, resource estimates, evidence gates, and handover records.
Boundary · The service structures an evaluation; it does not promise a useful quantum advantage or provider entitlement.
A separate security track
Post-quantum migration runs on classical systems today.
Learning circuits and migrating cryptography are different jobs. The security route starts with ownership and inventory—not with a QPU experiment.
Assign ownership and inventory
Connect cryptography to systems, data lifetimes, protocols, certificates, libraries, hardware, suppliers, and accountable owners.
Prioritize exposure and change difficulty
Rank long-lived data, externally exposed protocols, critical services, hard-to-update devices, and supplier dependencies.
Pilot standards-based replacements
Test ML-KEM for key establishment and ML-DSA or SLH-DSA for signatures where profiles and product support are appropriate.
Build crypto agility and migrate
Make algorithms replaceable while preserving security and operations; retain interoperability, rollback, performance, and audit evidence.
Answers and evidence
Open only what you need.
The core path ends above. These questions and sources are here when you need to verify a detail, provider boundary, roadmap item, or current standard.
Common questions
What is quantum computing in simple terms?
Quantum computing processes information with controllable quantum systems. A useful introduction follows how states are prepared, transformed by operations, and measured—not the misleading claim that a machine simply tries every answer at once.
Does quantum computing make every problem faster?
No. Quantum algorithms address particular problem structures under specific assumptions. Evaluation still needs a problem mapping, classical baseline, realistic resource estimate, and total-workflow evidence.
Do I need a physics degree or a quantum computer to start?
No. A careful circuit model, basic algebra and probability, and a normal computer are enough for the first local labs. Deeper hardware, algorithm, and error-correction work adds more physics and mathematics as the questions demand it.
What should my first quantum project be?
Build a Bell-state workflow: predict correlated outcomes, create H plus CX and measurements, run locally, inspect the source, vary shots, explain the distribution, and retain the evidence.
When should I run on real quantum hardware?
Use a QPU when physical target behavior can change the answer and the simulator workflow already passes. Record the target, ISA circuit, access, queue, usage, calibration context, job ID, evidence need, and stop condition.
Does QFlow Studio replace Qiskit or IBM Quantum?
No. Qiskit provides independent software tools, IBM controls its platform and hardware access, and QFlow Studio is a Neura Parse workflow, learning, and evidence layer around those kinds of tools and routes.
Is Neura Parse's IBM Partner Plus status an IBM Quantum endorsement?
No. It is a company-level ecosystem programme status. It does not imply IBM endorsement, IBM Quantum Network membership, a jointly developed or certified QFlow integration, bundled hardware access, privileged access, or research validation.
Is post-quantum cryptography the same as quantum computing?
No. PQC uses classical computers and quantum-resistant cryptographic standards. Learning quantum circuits and migrating enterprise cryptography are related strategic topics but different engineering programmes.
Primary sources
Last reviewed 11 August 2026. Check mutable software, access, roadmap, and standards details at the primary source before a course, pilot, or paid run.
IBM Quantum and QiskitCurrent SDK, local simulation, hardware workflow, compilation, provider, and roadmap context.
- Qiskit local quickstart
Reviewed 11 August 2026: the official local Bell quickstart recommends Qiskit 2.5.0-compatible packages and uses StatevectorSampler without a provider login.
- Exact simulation with Qiskit SDK primitives
Official boundary between local reference primitives and target-specific ISA preparation for Runtime workflows.
- Run a first circuit on IBM Quantum hardware
Official account, target selection, ISA transpilation, primitive execution, job, and analysis workflow.
- Qiskit transpiler documentation
Official current API context for rewriting circuits to target topology and supported instructions.
- IBM Quantum 2026 Technology Atlas
Updated March 2026. IBM labels roadmap items as current intent, goals, and objectives that can change or be withdrawn.
- IBM Partner Plus programme
Official programme overview. Neura Parse membership is company-level and does not establish product certification, IBM Quantum Network status, endorsement, or hardware entitlement.
NIST post-quantum standardsFinal standards, migration status, and current crypto-agility guidance for the separate PQC programme.
- NIST Post-Quantum Cryptography project
Updated 5 August 2026. NIST says organizations should begin applying the principal standards now and describes the transition toward 2035.
- FIPS 203 · ML-KEM
Final NIST module-lattice-based key-encapsulation mechanism standard, published 13 August 2024; its page carries a November 2025 potential-update notice.
- FIPS 204 · ML-DSA
Final NIST module-lattice-based digital signature standard, published 13 August 2024; NIST added a minor-errata planning note on 31 July 2026.
- FIPS 205 · SLH-DSA
Final NIST stateless hash-based digital signature standard, published 13 August 2024.
- NIST CSWP 39upd1 · Crypto agility
Final guidance updated 29 June 2026; it supersedes the withdrawn original CSWP 39 page.
- 2026 PQC updates to PIV working drafts
NIST's 12 June 2026 preliminary PIV materials illustrate incremental, dual-stack transition work; they are not final public drafts.
Neura Parse product and research recordsPrimary product documentation and the two public QANTIS paper records retained with their claim boundaries.
- QFlow Studio documentation
Product documentation for visual workflows, Academy, provider routes, runs, evidence, security, and APIs.
- QANTIS foundational paper · arXiv:2603.00785v1
Submitted 28 February 2026. The paper reports IBM Heron experiments and explicitly does not claim wall-clock advantage.
- QANTIS sequential belief-update paper · arXiv:2607.06760v1
Submitted 7 July 2026. The QPU is framed as a calibrated belief-update service inside a classical planning loop.
Start locally. Keep one clear question, one result, and one limitation together.