A Magic: the Gathering draft simulator that demonstrates game AI concepts
Role: AI Programmer
Developed in a month in December 2023
2-person team (Myself and Nico Toé)
Written entirely in C# and WinForms
Nico and I worked together to determine how the AI should make decisions. We ultimately decided on a utility-based AI that integrated some player-modeling aspects.
In addition to writing the architecture for how the draft functioned itself, I wrote the front-end code for how the user interacts with the draft simulator and the back-end code for how the AIs interact with the draft.
We wrote and presented our work and findings into a 13-minute presentation that details our methodology and thought processes behind the decisions that we made, as well as how to use the software we developed.
In addition to the presentation, we also wrote a 7-page paper detailing our work to a greater extent.
I have always been fascinated by the concept of making an AI that could play Magic: the Gathering effectively, and while this project is a far cry from making that a reality it's a fantastic step in the right direction! My teammate Nico Toè and I focused on modeling player behavior rather than finding optimal picks. There is a lot that we could do to improve upon this project in the future, but it was a lot of fun to develop and see emergent gameplay from our concepts!