As a small team of three, I was the sole programmer for the of Ludum Dare 56 games jam. This time around I was able to implement complex game design paradigms such as finite state machines and global event singletons. This game was designed with the theme: "Tiny Creatures" in mind. This game was met with a very positive reception with people appreciating the art style as well as the progression system that was implemented. In this game you are challenged to see how long you can survive, in a place that, depending on which region you're in, influences the likelihood of which enemy bugs you will encounter, you will be able to level up and evolve to receive different abilities to survive the odds, but your enemies also adapt and become stronger...

Participating in the global games jam 2024 with a team of developers, artists and designers from across the world including different states in USA, Italy, Puerto Rico and across the UK. Following the theme of "Make me laugh", we present: Morello's Mediocre night of Marvels, an ambitious platformer that requires finding depressed jesters and making them laugh, going through gags like the falling anvil to squish down to compact areas. I was a lot more experienced in Unity and programming concepts as well as handling version control to ensure a minimum viable product was achieved before the deadline of one week was due. I was able to assist my peers when roadblockers occured and was able to ascertain my own limitations when it came to developing the scope of this project.

Ruby's Adventure was my first experience using a game engine: Unity. Ruby's adventure is a Unity-provided tutorial on how to begin to design a small 2D game using sprites, animations and Object Oriented Programming with C#. I took this one step further, assigning different variations of NPCs, taking advantage of Unity's prefab system to create variations of enemies and characters that can do unique things such as changing dialogue when conditions are met and chaning movement patterns. Ruby must fix all of the robots in her village but perhaps someone else is pulling the strings. The game is available to play on Github and is the first major project I had taken outside of university.

Designing a fitness app from the ground up, utilising cloud computing to enhance the user experience with the Firebase API to store user login information securely. This app also uses mobile sensors such as the accelerometer to detect and count footsteps, to approximate burned calories for the sole purpose of aiding a healthier lifestyle in a clear, accessible interface, such features include using the MVVM model, Inversion of Control and tabbed pages for the UI.

Inspired by a friend and chef: Chad, a chatbot was developed using Discord's API and Edamam's recipe API that can recall any recipe in its database when asked for a recipe containing a key word, alternatively, random recipes can be selected and the nutritional values of the recipes can be indexed and provided, which is acheived using event callbacks. This was the first University project that I had done and was priveledged to work on this with others who also made significant contributions and this project introduced me to version control and programming in a team.

Using a dataset that provides information of the weather in Australia from Kaggle, the data was preprocessed via scrubbing of irrelevant and incomplete data, Encoded using OneHotEncoder to assign value to strings such as weather types which are non numerical. The model was trained and compared to a scaled size of the dataset, the more optimal of the two was chosen. Logistic Regression and K Nearest Neighbour (KNN) was used to predict if it would rain today. The models produced high levels of accuracy each, and were tuned to further increase predictibility using grid search and randomized search to find the optimal hyperparameters for Logistic Regression and the optimal number of neighbors for KNN. Confusion matrices and cross validation were performed to analyse the accuracy and precision of the training models. In conclusion: Both models succeed in classifying if there will be rain tommorow. The Logistic Regression model was the better model with a perfect accuracy when tuned whereas KNN had high accuracies of 89% when tuned