The article "Machine Learning Force Fields" by Oliver T. Unke et al. reviews the application of machine learning (ML) in computational chemistry, particularly focusing on the construction of ML-based force fields (FFs). The authors highlight how ML can bridge the gap between the accuracy of *ab initio* methods and the efficiency of classical FFs, enabling the study of large chemical systems with high accuracy. The key idea is to learn the statistical relationship between chemical structure and potential energy without relying on preconceived notions of fixed chemical bonds or specific interaction knowledge. The review covers the core concepts underlying ML-FFs, including the potential energy surface (PES), invariances of physical systems, and machine learning foundations. It also provides a step-by-step guide for constructing and testing ML-FFs, discusses best practices and common pitfalls, and lists several software packages that can aid in this process. The article concludes with a discussion of the challenges that need to be addressed for the next generation of ML-FFs to achieve broader applicability.The article "Machine Learning Force Fields" by Oliver T. Unke et al. reviews the application of machine learning (ML) in computational chemistry, particularly focusing on the construction of ML-based force fields (FFs). The authors highlight how ML can bridge the gap between the accuracy of *ab initio* methods and the efficiency of classical FFs, enabling the study of large chemical systems with high accuracy. The key idea is to learn the statistical relationship between chemical structure and potential energy without relying on preconceived notions of fixed chemical bonds or specific interaction knowledge. The review covers the core concepts underlying ML-FFs, including the potential energy surface (PES), invariances of physical systems, and machine learning foundations. It also provides a step-by-step guide for constructing and testing ML-FFs, discusses best practices and common pitfalls, and lists several software packages that can aid in this process. The article concludes with a discussion of the challenges that need to be addressed for the next generation of ML-FFs to achieve broader applicability.