Capstone Project — Quantum AI Specialist

Build a Hybrid Quantum AI Solution

Objective: Design, implement, and present a hybrid Quantum–Classical AI model that addresses a real-world, ethically-scoped problem. Deliverables include a runnable notebook, project report, and a short presentation. Successful completion awards the Quantum AI Specialist Certificate.

Estimated effort: 6–8 weeks (part-time)
Prerequisites: Completion of Modules 1–8 (foundational quantum computing, Quantum AI patterns, ethics, and hands-on labs).


What learners will accomplish

  • Build a full hybrid pipeline (data → encoding → quantum subroutine → classical model → evaluation).
  • Demonstrate practical choices for quantum components (PQC/VQE/QAOA/quantum kernels) and justify trade-offs.
  • Include an ethical risk assessment and deployment checklist.
  • Produce reproducible code and a professional presentation—suitable for portfolio or employer review.

Capstone timeline & milestones (6 weeks)

  • Week 0 — Topic selection & team formation (optional)
  • Week 1 — Project proposal (due Day 7)
    • 1-page problem statement, goals, dataset, chosen quantum approach, evaluation metrics.
  • Weeks 2–3 — Development sprint 1
    • Data processing, baseline classical model, prototype quantum module.
  • Week 4 — Development sprint 2
    • Integration, hyperparameter tuning, and initial evaluation.
  • Week 5 — Ethics & scalability review
    • Complete the ethics checklist, run lightweight scalability tests, document limitations.
  • Week 6 — Finalize and present
    • Final report (≤5 pages), runnable notebook, 5–10 slide presentation, optional demo video.

Milestone checkpoints: weekly mentor feedback and one formal mid-project review at end of Week 3.


Deliverables

  1. Project proposal (PDF/Markdown) — problem, goals, data sources, methods, and risks (max 1 page).
  2. Runnable Jupyter notebook — clear install/run instructions and reproducible outputs.
  3. Final report (≤5 pages) — background, methods, results, limitations, ethics & deployment checklist.
  4. Presentation (5–10 slides) — concise demo-ready deck.
  5. Optional: 3–5 minute demo video or recorded walk-through.

Submission format: GitHub repo or ZIP with notebook(s), report, slides, and a README containing reproduction steps.


Assessment & Grading Rubric

CriterionWeightDescription
Technical correctness & reproducibility40%Code runs end-to-end; results are replicable.
Empirical evaluation & analysis25%Sound metrics, baselines, ablation studies, and visualization.
Ethical & safety analysis15%Clear risk assessment, mitigation plan, and compliance considerations.
Novelty & design choices10%Innovative or well-justified hybrid approach.
Presentation & documentation10%Clear report, readable notebook, and concise slides.

Pass threshold: 70% overall and no critical ethics violations.
Resubmissions: Allowed once after mentor feedback for up to 2 weeks.


Recommended project archetypes & starter ideas

  • Quantum Drug Screening (Healthcare): Hybrid QSAR + VQE features to rank candidate molecules (small dataset).
  • Portfolio Optimizer (Finance): QUBO-based selection solved by QAOA or simulated annealing; compare to classical mean-variance.
  • Quantum NLP (Text): Use a PQC feature-transformer with sentence embeddings for classification or semantic search.
  • Logistics (Routing): Small VRP/TSP mapped to QUBO with hybrid decomposition and classical post-processing.
  • Climate Sampling: Quantum-assisted sampling for a toy stochastic model to estimate uncertainty in forecasts.

Mentorship & support structure

  • Mentor allocation: One mentor per project (or team) for weekly reviews.
  • Office hours: Twice-weekly drop-in sessions for troubleshooting and guidance.
  • Peer review: Mid-project peer feedback session; optional code review swap.

Ethics, IP & Publication policy

  • Ethics: Projects must include an ethics assessment describing dataset consent, potential harms, and mitigation measures. Projects that pose dual-use, biological risk, or security sensitivities will require additional review and may be restricted.
  • Intellectual Property (IP): Students retain IP by default, but the program may request a royalty-free license to showcase exemplary projects. If collaborators include external partners, clarify IP in the proposal.
  • Publication: Public dissemination is encouraged but must not violate safety rules; embargo options available for sensitive work subject to review.

Resources & templates

(Templates will be added to the course repo; mentors will provide links on Day 0.)


Certification & Badge

Certification criteria: Complete all modules, pass capstone evaluation (≥70%), and submit final artifacts to the course portal.


Showcase & Career support

  • Top projects spotlight: Selected projects featured on course site and newsletter.
  • Hiring pipeline: Option to share top projects with partner employers (with student consent).
  • Career guide: How to present quantum AI projects in portfolios and interviews (resume bullets, demo tips).

FAQ for Capstone Applicants

Q: Can I work in a team?
A: Yes — teams up to 3 are allowed. Each member should contribute and be listed with responsibilities.

Q: What compute is provided?
A: Access to cloud simulators and limited quantum hardware credits (if available). Additional paid credits are the student’s responsibility.

Q: How long does grading take?
A: Mentor feedback is weekly; formal grading is completed within 2 weeks of final submission.

Q: What if my project fails technically?
A: You can still pass if your report documents the failure, presents reasonable analysis, and shows learning — reproducibility and rigor matter more than positive results.


Next steps & CTA

  • Step 1: Draft and submit your one-page proposal by Day 7.
  • Step 2: Join the Capstone kickoff session (calendar invite).
  • Step 3: Clone the course repo and start from the Notebook template.

Ready to start? Click [Start Capstone — Submit Proposal] (course portal link) to begin.