Benjamin Lienhard

Scalable quantum computing depends on high-fidelity gate operations and efficient qubit readout. For gate control, analytically shaped GRAPE pulses enhance single-qubit robustness, suppressing amplitude drift errors by over an order of magnitude and reducing coherence-related gate errors by 40%. A sensitivity-adaptive closed-loop method enables 64-ns Fourier-series CZ gates with 99.9% fidelity, supported by signal-line distortion correction and adaptive cost functions. On the readout side, frequency-multiplexed measurement faces challenges from crosstalk and nonidealities. Machine learning-based signal discrimination, implemented on FPGAs, enables fast, low-error, real-time readout and reset. Together, these advances improve fidelity, reduce calibration overhead, and support the scalability of superconducting quantum processors.