This thesis explores innovative computational design methods to revitalize adaptive vernacular architectures in rural Chinese settings, addressing the challenges posed by the rapid homogenization and cultural erosion resulting from modern construction practices. The study introduces two bespoke algorithms, GrowFill and TopoWood, tailored to optimize structures using local materials—earth and wood—commonly found in these environments. These algorithms are synergistically integrated, forming a multi-material, performance-driven system that not only enhances structural integrity but also improves material efficiency. GrowFill is designed for optimizing earth walls with a focus on advanced 3D printing techniques that adapt to structural needs and minimize material waste. TopoWood, on the other hand, reinterprets traditional timber frameworks, optimizing the use of wood through a structural performance-driven design that aligns with stress lines, thus facilitating less material use and greater spanning capabilities. Both systems utilize topology optimization to dictate the most efficient material distribution and structural form, proving effective in practical applications, as demonstrated through prototype developments by PLA and Clay printing tests. This design methodology not only counters the loss of architectural diversity and cultural identity but also proposes a sustainable model for future rural constructions by reducing environmental impact and supporting local craftsmanship. The thesis acknowledges the limitations inherent in the current scope of the algorithms and suggests further research to expand their application to three-dimensional optimization and larger-scale implementations.