Inspiration
I love to play board games, but I often can't get a big enough group together to play. This led to my original idea: an AI-powered opponent in Tabletop Simulator who could play virtual board games against you after reading their rules books. This proved to be too ambitious, so I settled on a simplified case for my project: a bot for a modified version of the card game Cheat, built in Python.
Features
The project features a graphical implementation of Cheat in Pygame and a bot integrated with the OpenAI API that plays against the user.
Challenges
Towards the end of the hackathon, I had some struggles integrating the OpenAI API and the features it brought into the Pygame piece of my project, but I ultimately found a solution and got the bot working.
Future Plans
I feel this project is still incomplete, that the bot could be improved to have a longer memory, or that maybe it would be better if I tried a reinforcement learning approach instead. Nevertheless, I enjoyed learning to work with APIs and hope to continue learning with this project, however that may be!
Log in or sign up for Devpost to join the conversation.