Many bioinformatics tools today expose a range of user-controllable parameters, tied to the algorithm, that will affect the outcome of the run. The complexity is increased as softwares are tied together in pipelines. Only using the default parameter settings are not guaranteed to yield the optimal outcome. Instead, in this project we want to implement design of experiments methodology for finding optimal software parameter settings in software pipelines. The implementation will investigate a search space, calculate a model from the result, and predict the optimal parameters. If the optimum is predicted outside the current search space, the space will be moved in the direction of the optimum. This process is repeated in an iterative manner until no further improvement can be achieved. There is also an option to start with a screening round, using the recently developed generalized subset design (GSD). The aim is to present a user-friendly python package that allows users to optimize their software pipelines.