In this study, we conducted calculations to determine the thermal conductivity of silicon nanowires with various shapes and cross-sectional designs using the Monte Carlo method combined with Green–Kubo heat flow autocorrelation. This computational approach is known for its reliability in predicting thermal properties of complex nanostructured devices. We specifically examined nanowires with circular or rectangular sections combined to complex serpentine and fishbone modulations. Our findings indicate that the nanowire's geometry significantly impacts the phonon mean free path and thermal transport. More specifically, patterns with serpentine modulations exhibit a pronounced reduction in thermal conductivity attributable to enhanced phonon boundary scatterings. Our simulation results are compared with the available experimental data, highlighting the potential of our method for structural optimization in thermal management applications, particularly in devices like thermoelectric converters.