2026 Internship Project: Hamiltonian Learning

As an intern, you will work on implementing algorithms to learn the native Hamiltonian and noise governing the qubits on a superconducting quantum chip during parallel gate execution. You will model the effective Hamiltonian and noise on a small multi-qubit subsystem, co-design experiments to validate the model, reconstruct the Hamiltonian and noise from numerical simulations, and validate the model on the chip. You will produce a pre-production research codebase which can be used for continued development to learn the Hamiltonian and noise. This is a rare opportunity for an intern to work on cutting edge research outside academia, with the opportunity to interface with benchmarking, calibration, and device theory teams.

Additional points: There is flexibility to work remotely or in the office at Berkeley. Your mentors will be remote. The internship should last 3 months over the summer of 2026, with flexible start and end dates.