This course is popular
3 people enrolled last week.
Custom optimizer for QAOA to run on real quantum hardware, not just a simulator
‼️Till August 22, 50% off (code Early) and 80% off for developing nations (code Dev).
Quantum optimization is the most commercially scrutinized corner of quantum computing, yet most engineers who want to assess it face a gap: tutorials are either toy demos with no real data or papers that assume a physics degree. This workshop closes that gap in one sitting. In 2.5 hours, you will derive QAOA from first principles, build the complete pipeline on real market data (10 and 6 US stocks, actual return and covariance figures), run it in Qiskit 2.4, and benchmark it against the exact classical optimum.
What you will get:
⭐ Full QAOA in 2.5 hours — theoretical foundations and Qiskit codes
⭐ Four Jupyter notebooks — Yahoo data fetching, Qiskit Primitives, QAOA on simulator for 10 stocks, and on IBM hardware for six stocks
⭐ IBM Hardware — Custom optimization routine to run on hardware in a series of single jobs through open plan
⭐ All PDF slides — complete derivations written to be a standalone reference
⭐ Full recording — Full recording available in two hours
⚛️ "Do I need a paid IBM account?" No. The custom optimizer has been specifically developed to run on the free IBM Open Plan in 2 minutes.
What you’ll learn
Master the full QAOA pipeline on real market data — and become the person your team trusts to assess quantum optimization.
Cast Markowitz portfolio selection — return, risk, and a budget constraint — as a QUBO
Set penalty weights that actually work (too small: infeasible portfolios win; too large: the landscape flattens)
Map any QUBO to an Ising Hamiltonian via xᵢ → (I − Zᵢ)/2 and verify the mapping numerically
Derive the ansatz from the adiabatic theorem through Trotterization to learnable parameters
Translate every covariance entry into an explicit CNOT–Rz–CNOT circuit block
Visualize the full (γ, β) optimization landscape at p = 1 — the picture that makes QAOA click
Build the ansatz with QAOAAnsatz, estimate ⟨H_C⟩ with EstimatorV2, sample with SamplerV2
Train with COBYLA using adiabatic-inspired linear-ramp initialization and multi-restart
Decode measured bitstrings into named portfolios and filter for budget feasibility
Compute approximation ratio and probability-of-optimum against exact brute-force enumeration
Transpile for a real utility-scale IBM device and read the depth and gate-count cost of dense covariance matrices
As of 2026, QAOA has no demonstrated advantage over classical solvers at accessible problem sizes. This workshop teaches you why.
Workshop agenda
0:00–0:20 | The Problem, Classically
Set up real market data (10 US stocks: Apple, Microsoft, JPMorgan, etc., with actual returns and covariances from Yahoo Finance). Frame portfolio selection as maximizing return and minimizing risk
0:20-0:50 | Flash Introduction to Quantum Computing and Qiskit
Learn the basic quantum gates and how to implement circuits along with measurements using Qiskit primitives.
0:50–1:15 | From QUBO to the Ising Hamiltonian
Map the classical cost function to the quantum Hamiltonian via the eigenvalues of quantum gates
1:25–1:50 | QAOA Mechanics
Motivate QAOA from the adiabatic theorem: start in the easy ground state of a mixer, evolve under an interpolation between the mixer and the cost Hamiltonian, and discretize into alternating unitaries
1:50–2:15 | Hands-On Coding: Five Checkpoints
Implement QAOA in Qiskit by dividing it into five checkpoints for 10 stocks using the Aer Simulator. Each checkpoint is self-contained; if you fall behind, jump to the next checkpoint marker.
2:15–2:30 | Custom Optimization Loop for Hardware Run
Revisit the complete pipeline and add Checkpoint 6 after running on IBM hardware for 6 stocks.
What you take home
Four Jupyter notebooks (fetching live stock data & classical optimal, Qiskit Primitives, QAOA on 10 stocks with Aer Simulator, QAOA on 6 stocks with IBM hardware using Free Plan). All slides.
Learn directly from Muhammad
Dr. Muhammad Faryad
Tier-2 IBM Qiskit Advocate, IBM-certified Qiskit developer
Who this workshop is for
ML engineer at a bank tasked with evaluating quantum optimization — strong ML background, no QC experience, want your own honest benchmark.
Physicist or QC researcher who knows the theory — want to see a complete, honestly-benchmarked real-world QML application, end-to-end.
Risk manager or CTO asked to evaluate quantum computing for your firm — technical enough to judge, tired of slick marketing and hype.
What's included
Live sessions
Learn directly from Dr. Muhammad Faryad in a real-time, interactive format.
Slides, Jupyter Notebooks, and recording
All material used during the lecture will be provided to the students ONE DAY BEFORE the workshop begins so that they can follow along. The recording will be provided 2 hours after the workshop ends.
Lifetime access
Go back to course content and recordings whenever you need to.
Community of peers
Stay accountable and share insights with like-minded professionals.
Certificate of completion
Share your new skills with your employer or on LinkedIn.
Maven Guarantee
Your purchase is backed by the Maven Guarantee.
Free resource
Five Jupyter Notebooks showing classical data encoding is the permanent bottleneck of Quantum ML
Frequently asked questions
Maven for Teams
Reimbursement
Get your company to pay
Everything L&D needs: email template, receipts, and certificate of completion.
Get reimbursedTeam discount
Learn with your teammates
Save 20%+ when 2 or more teammates enroll in the same cohort.
Save 20%+ with a teamPrivate cohort
Run a cohort for your org
A dedicated cohort with a custom schedule and curriculum, tailored to your team.
Book a private cohort$200
USD
8 days left to enroll
11am–1:30pm EDT
·1 more cohort