AWS Quantum Practitioner - 6 Week Practice Plan¶

This plan assumes 10 hours per week. Quantum has a steeper conceptual ramp than other AWS specialties; budget more time if linear algebra is rusty.

Week 1: Math and Quantum Foundations¶

Goals: rebuild the math, internalize qubits.

• Refresh complex numbers, vector inner product, matrix multiplication
• Read Nielsen & Chuang chapter 2 sections on state vectors and unitary evolution
• Learn Dirac notation: |0>, |1>, |+>, |->, ket/bra
• Understand superposition: |psi> = a|0> + b|1>, |a|^2 + |b|^2 = 1
• Understand measurement: probability = |amplitude|^2
• Visualize on the Bloch sphere
• Learn Pauli matrices, Hadamard, phase gates by hand on paper

Deliverable: by hand, compute the result of H|0>, X|0>, Z|+>.

Week 2: Circuits and Gates¶

Goals: read and write quantum circuits.

• Single-qubit gates: I, X, Y, Z, H, S, T, Rx(theta), Ry(theta), Rz(theta)
• Two-qubit gates: CNOT, CZ, SWAP
• Three-qubit gates: Toffoli (CCNOT), Fredkin (CSWAP)
• Bell state preparation: H on qubit 0, then CNOT(0, 1)
• GHZ state preparation
• Quantum teleportation circuit (canonical exercise)
• Install the Braket SDK, run a Hello Bell on the local simulator

Deliverable: write a Bell-state circuit and a 3-qubit GHZ in Braket Python.

Week 3: Amazon Braket Hands-On¶

Goals: get fluent with the Braket service.

• Open the Braket console; tour notebooks, devices, tasks, hybrid jobs
• Run a circuit on LocalSimulator
• Run the same circuit on SV1
• Run on a real QPU (IonQ Aria, Rigetti Ankaa, or IQM Garnet)
• Be deliberate; even small jobs cost a few dollars
• Inspect results stored in S3
• Set IAM permissions for braket actions
• Set up cost budgets and tagging on Braket tasks

Deliverable: a working notebook that runs a Bell circuit on a simulator and a QPU and compares results.

Week 4: Hybrid Workflows and PennyLane¶

Goals: build variational algorithms.

• Read Braket Hybrid Jobs documentation
• Run a managed PennyLane container hybrid job
• Implement VQE for H2 molecule (Braket Examples has a notebook)
• Implement QAOA for Max-Cut on a small graph
• Understand priority queueing and reservation
• Compare simulator vs QPU quality for the same VQE

Deliverable: a working VQE that finds the H2 ground state energy on simulator.

Week 5: Algorithms and Applications¶

Goals: know the canonical algorithms.

• Grover's algorithm: implement and run; understand quadratic speedup
• Quantum Phase Estimation: implement on simulator
• QFT (Quantum Fourier Transform): implement and verify
• Read about Shor's algorithm (conceptual; do not implement at scale)
• Read about quantum machine learning (kernel methods, variational classifiers)
• Read about analog Hamiltonian simulation on QuEra Aquila

Deliverable: notebook implementing Grover for 4-qubit search.

Week 6: Operations, Cost, Mocks¶

Goals: pass under realistic conditions.

• Master Braket pricing model: per-task, per-shot, device fees, simulator per-minute, hybrid job compute
• Practice cost estimation for circuits
• Review error mitigation: zero-noise extrapolation, readout error mitigation
• Review noise model differences: trapped ion vs superconducting vs neutral atom
• Two timed mock exams using exam-style questions
• Review weak areas

Daily Cadence¶

• 30 minutes math / theory
• 60 minutes hands-on Braket
• 30 minutes algorithm reading or scenario practice

Lab Costs¶

• Local simulator: free
• SV1, DM1, TN1: cheap, often pennies per minute
• IonQ Aria task: typically a few cents per shot, plus per-task fee
• Rigetti, IQM: similar order of magnitude
• Hybrid Jobs: classical compute fee per minute plus quantum task fees

Budget 50-150 USD for hands-on practice across the plan.

Red Flags You Are Not Ready¶

• Cannot compute the matrix for H or CNOT from memory
• Cannot prepare a Bell state in Braket
• Do not know the difference between SV1, DM1, TN1
• Cannot explain when to use a QPU vs a simulator
• Cannot estimate the cost of a 1000-shot circuit on IonQ Aria
• Mock exam scores below 65 percent