Welcome to Nicholas Steiner's Portfolio
I am a senior at Oakland University studying Computer Science looking to continue my career in the Data Science field.
As a Data Science Intern at Lincode Labs, Inc. I've designed and implemented computer vision solutions for many customers. With my passions for coding and problem solving I've created many personal and collaborative projects that you can see down below.
My Portfolio
Food Junkie
A collaborative Android application developed with 5 others to enable storage and creation of user recipes, track ingredients users have, and provide other useful cooking resources.
My main contribution to Food Junkie was creating the database to store user recipes and ingredients and incorporate the access of information with front-end designs through SQL queries. I utilized database design and implementation techniques to efficiently gather data to present to the user with minimal latency.
As the lead developer I assisted in debugging many features of the application, including a front-end design flaw that did not allow scaling of the application across different devices. I also did much of the project management, coordinating feature creation and scaling the project scope across different development phases.
Space Frontier
January 2023 - April 2023
Healthy Foods Classifier
A 2D space shooter game independently developed utilizing C++ and the Simple Fast Multimedia Library. In Space Frontier, users control their ship and try to shoot enemies while dodging incoming attacks, gaining power ups along the way. Players are also able to keep track of their high score which is kept in a database and accessed with SQL.
This project required the use of Object-Oriented Programming and OOP principles throughout development. Players encounter different types of enemies created through inheritance, separate behavior of these enemies, the player, and power-ups are handled through encapsulation, and the shooting mechanisms seen from the player's and enemy ships is controlled through polymorphism.
July 2023 - August 2023
An ML project that aims to find if a classical machine learning model can perform better at classifying foods as healthy or unhealthy than a CNN achieving 90.37% accuracy.
The dataset I created for this project contains 221 different foods with information from their nutrition facts and a label: healthy, unhealthy, or an added third class, should be eaten in moderation. After data normalization, augmentation, and model comparison I built a K-Nearest Neighbors model that was able to achieve 90.76% accuracy with the addition of a third class.
October 2024 - December 2024
