AI2 Tango is a platform that allows you to build machine learning experiments out of steps that can be reused and repeated. It eliminates duplicate work, stores past results, and keeps track of how they were obtained so you can quickly identify and trace back your best work. Experiments are flexible, so you can change direction, pursue another idea and incorporate feedback, without losing previous results. Write your own steps by wrapping common Python functions, or take advantage of the library of pre-built steps for training models, working with datasets, and running evaluations. It is integrated with popular tools like the Huggingface transformers library, PyTorch Lightning, and others.