We present a new game, “MLA: Machine Learning Arena”, in which the player’s goal is to train a machine learning agent to win a boxing match. The game features multiple phases, fully animated characters for the player to control, and machine learning integration. We tackled several technical and design challenges in this MQP, including: 1) Communicating machine learning progress through UI elements to the user, 2) Collecting poor training examples from users, 3) How to introduce the ideas behind machine learning to players who have no background in machine learning. We conduct user testing with 27 human participants to determine if players feel that the ML is learning from them. We found that 85 percent of players were able to tell the difference between a random agent and the trained agent. Based on our development experience, we offer suggestions for future development: 1) Continue the research on communicating machine learning to players through either gameplay or UI metrics, 2) Use input sanitization to make the machine learning feel more fulfilling to players, 3) Plan far ahead of time for a lot of extra development time for machine learning implementation.