Worked on solver integration and optimization features for open-energy-transition/pypsa-eur and cvxgrp/cvxpy-ipopt, focusing on large-scale linear programming in Python. Delivered configuration-driven support for the Xpress solver, enabling both default and GPU-accelerated solving with barrier method settings to improve scalability and performance for energy system modeling. Enhanced the XPRESS solver interface for compatibility with Xpress 9.8, adding bounded variable support while maintaining API stability. Emphasized reproducibility and flexibility in scenario analysis workflows, using Python and YAML for configuration and documentation. The work demonstrated depth in solver development, integration, and optimization, addressing both performance and modeling expressiveness.