Self-directed STEM AI Evaluation
Conducted self-directed evaluation of AI-generated STEM responses in a research context. The work involved reviewing physics and mathematics solutions, checking derivations, identifying conceptual and algebraic errors, verifying assumptions and boundary conditions, and assessing whether scientific Python/Jupyter code correctly implemented the intended physical models. The evaluation focused on quantum mechanics, condensed matter theory, non-Hermitian physics, matrix eigenvalue problems, perturbation theory, numerical simulations, and visualization.