Computational physics courses are becoming an integral part of the undergraduate physics curriculum because they prepare students for their post-baccalaureate career where computational skills are applicable to research and industry jobs. The instruction of computational physics courses varies greatly based on curricular requirements and departmental resources. At Oregon State University, the computational courses are a project-driven laboratory experience. The computational physics lab course is strongly related to the content delivered in the junior-level lecture courses, but they are taught by different instructors so that a non-computational physicist can still deliver the lecture material. In the computational physics lab course, students solve physics problems related to lecture course content using Python and pair programming, a coding technique shown to improve code quality and learning outcomes. We will present the design choices of using the lab structure and the coordination with lecture courses, and the development of the computational physics lab course.