I'm a Software Engineering student, and this is where I share my projects. Here you'll mostly find my university work, all of great value, along with some smaller personal projects.

• Name: Nicolas Debacher
• Course: Software Engineering
• Skills: Java, C#, ASP.NET Core, Spring Boot, JPA Hibernate, PostgreSQL (PL/pgSQL)
• Interests: Software architecture and performance, design patterns
• Currently learning: Rust

I have a solid foundation in logic programming and currently tutor in the Distributed Systems course at my university. I enjoy exploring new technologies and finding efficient ways to solve problems.

A web application built with Spring and React for a university integrative project with three other colleagues. We received a ready-made requirements documentation and had to build the designed application. The application predicts energy costs based on the usage times of household appliances registered by the user. We used some external APIs to get tariff flag values and reports from energy concessionaires.

A project created for a technical hiring challenge. It's an MVC application with Spring Boot and Java Swing that queries the ViaCep API to get details about a city.

A project for a Distributed Systems course in the 6th semester. The context was to develop a traffic simulation on a road network represented by a numerical matrix. The goal was to apply mutual exclusion rules and prevent deadlock and starvation. Each car had to follow the other, occupying each cell of the matrix individually until reaching the end of the road. Intersections required cars to decide the path to take to gain possession of all necessary cells, allowing one car to pass at a time. Check out the GIF below for a clearer view of how it works:

An ASP.NET Core API designed to serve a music streaming app. The app, Yota, was developed for the Mobile Development college subject using .NET MAUI. Due to time constraints, the API wasn't used in the final project, and we prioritized using Deezer's API for convenience. However, it still serves as a reference for creating a CRUD scenario with the following model:

A university project where we created a linear programming solution for the motifs problem. The goal was to find motifs with the highest number of possible connections. A motif is a set of same-colored nodes in a graph. We generated random graphs with different colors and connections and searched for subgraphs with the highest number of possible connections using breadth-first and depth-first search. We used Python to handle data structures more easily.

The latest integrative project. We received a dataset with approximately 800 data points of various fruit species and trained two different algorithms: Random Forest and Decision Tree. After training, we created a REST API with one endpoint that receives a CSV of some fruits and returns them classified. Users can choose which model to use for classification. We gathered data on the performance of the different models and obtained the following confusion matrices:

From this, we concluded that the Random Forest model, despite being slower, has a higher accuracy rate compared to the Decision Tree model.

I'm working on other personal projects and plan to keep updating this space. If you have any questions about the repositories, feel free to email me at ndebacher@gmail.com.