Learned and became proficient in the creative coding library p5.js. Incorporated physics properties into game design, such as collusion between objects and gravity, making games more realistic and increasing player engagement. Led four team members to create three unique web games, including avatar design, CYOA, and collection style.

Trained a GAN model to recolor historical images and analyzed biases within its color choices. Tested several methods, including Hard De-biasing, WEAT, and ECT, to de-bias word embedding datasets. Assessed bias within the COMPAS, an algorithm used in U.S. courts to predict the likelihood of a defendant to re-offend, with a SVM model.

Trained a CNN model to detect lung nodules from medical-used images. Applied machine learning in humanitarian aid, estimating displacement patterns of Somalians due to extreme weather. Predicted refugee relocation patterns in North Africa based on VHI with Conv1D model. Developed a basic medical chatbot that detects signs of depression in messages.