Examples — bring-your-own starting points

62 notebooks lifted from the upstream uniqx gallery. Use any of them as scaffolding when designing a custom workload for the “bring-your-own” track.

Every notebook follows the same skeleton: problem definition → trace with uniqx → preflight() → submit to whichever route the engine recommends → compare to a classical oracle. The user code is identical regardless of the route the engine picks — that is the hardware-agnostic property the hackathon scores you on.

Foundational — read these first

Notebook	What it teaches
`getting_started.ipynb`	Vector add, matmul, eigs. Trace + submit + parse round-trip.
`hybrid_cpu_gpu_qpu.ipynb`	The hackathon’s central theme — same code, three hardware routes, `preflight()` shows the tradeoff.
`hardware_aware_dialects.ipynb`	How the lowering pipeline decides what runs where.
`benchmark_chemistry_routes.ipynb`	Benchmark the same chemistry workload across every available route.

Algorithm primers

Notebook	Algorithm
`algorithm_grover_primer.ipynb`	Grover amplitude amplification
`algorithm_qae_primer.ipynb`	Quantum amplitude estimation
`algorithm_qite_primer.ipynb`	Quantum imaginary-time evolution
`algorithm_hybrid_quantization_primer.ipynb`	Hybrid classical/quantum quantization

Chemistry — DFT track and beyond

Notebook	Problem	Classical oracle
`chemistry_ground_state.ipynb`	H₂ ground-state energy	Exact diagonalization
`chemistry_hamiltonian_representations.ipynb`	Build / inspect molecular Hamiltonians	PySCF
`vqe_ground_state.ipynb`	Variational ground state	eigsh
`two_electron_chemistry_vqe.ipynb`	Two-electron VQE end-to-job	FCI
`real_space_quantum_chemistry.ipynb`	Real-space basis instead of Gaussians	PySCF
`nmr_notebook.ipynb`	Isotropic shielding tensors	PySCF / NMR prop
`geometry_optimization.ipynb`	Equilibrium geometry via gradients	PySCF
`ligand_geometry_optimization.ipynb`	Ligand pose / strain optimisation	UFF
`conformer_search.ipynb`	Conformer enumeration	RDKit
`transition_state.ipynb`	Saddle-point search	dimer / eigenvector following
`neb_reaction_path.ipynb`	Nudged elastic band	classical NEB
`photoisomerization.ipynb`	Excited-state dynamics	TDDFT
`allosteric_simulation.ipynb`	Protein allosteric coupling	MD biophysics

Physics — CFD, MD, spin systems

Notebook	Problem	Classical oracle
`aerodynamic_modeling.ipynb`	Aerodynamic flow modelling	NumPy CFD
`spin_chain_ground_state.ipynb`	TFI lowest eigenvalue	Lanczos
`spin_chain_dynamics.ipynb`	e^{-iHt}·ψ	scipy.expm
`large_spin_chain_dynamics.ipynb`	Larger-N spin chain	block expm
`quantum_simulation.ipynb`	Generic Hamiltonian simulation	Trotter
`poisson_solve_grid.ipynb`	Lu = b on a 2D grid	LU
`kinetic_eigenmodes_grid.ipynb`	Eigenmodes of ∇²	Lanczos

Numerical linear algebra and graphs

Notebook	Problem	Classical oracle
`least_squares_regression.ipynb`	lstsq(X, y)	Normal equations
`low_rank_denoising.ipynb`	Truncated-SVD denoising	LAPACK
`pagerank_dominant_eigenvector.ipynb`	PageRank	Power iteration
`graph_spectral_clustering.ipynb`	Spectral clustering	eigh + KMeans
`partition_function_logsumexp.ipynb`	logsumexp Boltzmann sampling	scipy

Sampling and statistics

Notebook	Problem	Classical oracle
`markov_chain_mixing.ipynb`	MCMC mixing rates	Power method
`variational_monte_carlo.ipynb`	VMC sampling	PRNG
`thermal_state_sampling.ipynb`	Thermal-state samples	classical MCMC
`random_walk_search.ipynb`	Random-walk search	classical walk
`mcmc_cpu_vs_gpu.ipynb`	Direct CPU vs GPU sampling — a model for reporting a hardware tradeoff.

Machine learning

Notebook	Problem	Classical oracle
`neural_network_training.ipynb`	Gradient-based training step	NumPy backprop
`dense_neural_network_hybrid.ipynb`	Hybrid dense MLP	classical MLP
`generative_adversarial_step.ipynb`	GAN training step	classical GAN
`reinforcement_learning_step.ipynb`	RL action sampling	tabular Q-learning
`binary_classification_quantum.ipynb`	Binary classifier	logistic regression
`kernel_svm_quantum_feature_map.ipynb`	Kernel SVM	RBF kernel
`vqc_softmax_readout.ipynb`	Variational quantum classifier	logistic
`qml_loss_reduction.ipynb`	QML loss reduction (QAE-as-mean)	numpy reduce
`gradient_variance_diagnostic.ipynb`	Variance of gradients (barren-plateau diag)	numerical gradient

Optimisation

Notebook	Problem	Classical oracle
`qaoa_maxcut.ipynb`	MaxCut on small graphs	Simulated annealing
`qaoa_with_layer_norm.ipynb`	QAOA + LayerNorm trick	classical SA
`combinatorial_qubo_optimization.ipynb`	QUBO solver	tabu / simulated annealing
`route_optimization.ipynb`	Vehicle routing	OR-Tools
`grid_expansion_planning.ipynb`	Energy-grid expansion	MILP

Quantum + classical interop

Notebook	What it shows
`classical_quantum_interaction.ipynb`	Round-tripping data between classical and quantum kernels.
`classical_quantum_roundtrip.ipynb`	Submit-and-resubmit pattern for hybrid loops.
`constrained_param_unitary.ipynb`	Parameter constraints on a unitary.
`jax_gap_primitives.ipynb`	JAX-style differentiable primitives mapped to gateway ops.
`oqi_usecases.ipynb`	Selected industrial use cases driven by `oqi` primitives.

Real-world demonstrators

Notebook	What it shows
`autonomous_driving_vla.ipynb`	Visual-language-action model with hybrid inference.
`fraud_detection.ipynb`	Imbalanced classification + quantum kernel feature map.
`threat_detection.ipynb`	Anomaly detection with a hybrid scorer.
`quantum_cryptography.ipynb`	QKD-style protocol primitive.

How to use these in a “bring your own” submission

Pick the notebook closest to the workload you want to build.
Copy it into submissions/<team-handle>/submission.ipynb.
Replace the problem definition with yours. Keep the preflight() → submit() → oracle-compare skeleton.
Fill in results.json.workload_description (required for track: "custom").
Submit per docs/submission.md.

The judges score you on the shape of your Pareto frontier and the quality of your justification, not on which example you started from.