Individuals seeking treatment for mental health problems often have to choose between several different treatment options. For disorders like depression and PTSD, many of the available treatments have been found to be, on average, equally effective. Research on precision medicine aims to identify the most effective treatment for each patient. This work is based on the idea that individuals respond differently to treatment, and that these differences can be studied and characterized. The push for personalized and precision approaches in mental health involves identifying moderators - variables that predict differential response into treatment recommendations. Unfortunately, there has been little real-world application of these findings, in part due to the lack of systems suited to translating the information in actionable recommendations. This dissertation will review the history of treatment selection in mental health, and will present specific examples of treatment selection models in depression and PTSD. Differences between treatment selection in the context of two equivalently effective interventions and stratified medicine applications in which goal is to optimize the allocation of stronger and weaker interventions will be discussed. Methodological challenges in building (e.g., variable selection) and evaluating (e.g., cross-validation) treatment selection systems will be explored. Approaches to precision medicine being used by different groups will be compared. Finally, recommendations for future directions will be made.