Graduate computational chemistry training and portfolio simulations for AI evaluation (Python + ORCA/GROMACS-style workflows)
Designed verifiable computational chemistry problem suites intended for AI evaluation using deterministic simulation outputs. Built Python automation to generate inputs, execute external simulation pipelines, parse results, and validate numerical stability. Structured tasks around quantum chemistry, molecular dynamics, and docking so model outputs can be compared against ground-truth metrics. • Quantum chemistry energy extraction pipeline with automated geometry generation and batch execution • MD workflow structuring and trajectory analytics (RDF, MSD, diffusion estimation) • Docking/binding affinity parsing and ranked-conformation evaluation • SCF convergence and numerical stability analysis detecting oscillations, drift, and non-convergence regimes