AlphaFold2 is an open-source artificial intelligence program that predicts the 3D structure of proteins based on their amino acid sequence. The program and its underlying algorithm solve a long-standing problem in structural biology, so adoption by the research community has been swift and enthusiastic. Running AlphaFold2 involves a simple input file and few parameter specifications, but requires access to large-scale computational resources such as GPUs and enough RAM to perform a search against a collection of datasets exceeding 2 TB. This session, aimed at researchers and research facilitators, will provide an overview of running AlphaFold2 on Alpine, covering performance considerations and limitations, predicting monomer and multimer proteins, checking AlphaFold2 outputs, and how to take advantage of spin-offs like ParaFold. Participants will receive training accounts for Alpine and can follow along using a hands-on example.