Developed a new feature for the lab-cosmo/atomistic-cookbook repository that enables calculation of silicon lattice thermal conductivity using the kaldo package and UPET machine learning potential. The work focused on implementing a reproducible workflow based on the Boltzmann transport equation, providing a complete recipe with detailed documentation and runnable Python examples. This addition supports materials researchers in performing advanced thermal transport analysis and enhances the cookbook’s scientific computing capabilities. The project involved close collaboration with other contributors and leveraged skills in Python, data analysis, and machine learning to deliver a robust, user-friendly solution for computational materials science.