Used the FastAI library to classify microscopic fungi images from the DeFungi dataset. This was done using transfer learning on ResNet and Vision Transformer (ViT) models. Achieved an overall accuracy of ~93% compared to ~85% from the original paper.

An interactive Connect Four game with a powerful AI opponent. The AI uses a depth-limited minimax algorithm with alpha-beta pruning. It was written in C++ and compiled to WebAssembly resulting in a blazingly fast AI that can run locally on the browser. The interface was made using HTML, CSS and JavaScript.

Designed and developed an iOS app that uses statistical insights and data visualizations to aid users in their weight loss journey. Built using SwiftUI, Swift Charts and CoreData. Published to the AppStore on May 2023.

An educational puzzle game about cryptography, for iPadOS. Decode a cryptic message by using various cryptography techniques that were used throughout history. This app was my submission to the WWDC23 Swift Student Challenge for which I won a scholarship from Apple.