Our Research

Quantum Hybrid Modules for AI: Attention, Optimization, and Verification on Near-Term Quantum Hardware

Vikram Lex

A modular framework for inserting quantum subroutines into classical AI pipelines. Comprises QSANN (quantum self-attention via Hadamard-test inner products), QAOA for combinatorial optimization, and Grover-accelerated neurosymbolic verification. Validated through 22 experiments spanning statevector simulation, cloud simulators, and execution on Rigetti Cepheus (8–100 qubits) and IonQ Forte (up to 30 qubits) QPUs. We do not claim near-term quantum advantage; we characterize the gap between asymptotic theory and present-day hardware.