Kenyon Leblanc
Software Engineer @ Cinteot - Undergraduate
I have a passion for technology and am always eager to experiment with new tools and dive into the latest advancements in the field.
My interest particularly lies in applications of AI and computer vision in image recognition and cybersecurity, where I see immense room for innovation.
I find the potential of AI and computer vision exciting as they have the capacity to transform the way we engage with our surroundings.
While I am excited by these technologies, I am also aware of the potential ethical implications they bring, emphasizing the need to stay informed and engaged.
Shippensburg University
Certifications
Below are some of my skills, and I'm always looking to learn more.
Having four years of experience with C, I have developed a deep understanding and proficiency in this foundational language. My journey involved the creation of numerous complex projects such as a Shell, a web server, and a compression program. I also leveraged C extensively in my operating systems class, further solidifying my comprehension and skills.
With three years of experience in Java, I've not only utilized it in various projects but have also gained an in-depth understanding of Design Patterns and Object-Oriented Programming (OOP). My knowledge of Java has enabled me to develop robust, scalable, and efficient applications.
I've been using Python extensively for the past two years, particularly in web development and AI projects at Cinteot. My Python expertise is augmented by my proficiency in Django, a high-level Python web framework that encourages rapid development and clean, pragmatic design.
My two-year experience in web development at Cinteot and personal projects has honed my skills in HTML, CSS, and JavaScript. I have used these technologies to build user-friendly, responsive, and interactive websites.
Bootstrap, a popular HTML, CSS, and JS framework for developing responsive, mobile-first projects, has been an essential tool in my web development journey. I have effectively incorporated Bootstrap in this personal website and several other projects, leading to highly responsive and visually appealing interfaces.
At Cinteot, I've extensively used Django, a high-level Python web framework that facilitates rapid development and clean, maintainable architecture. Django follows the DRY (Don't Repeat Yourself) principle, which makes it a highly efficient tool for web development.
My work at Cinteot has also involved Sentiment Analysis, an area of Natural Language Processing that I find fascinating. The process of understanding, extracting, and processing subjective information from source materials is a powerful tool when combined with AI capabilities.
At Cinteot, I've used TensorFlow, a powerful open-source software library for machine learning, to develop a model used in a Predictive Analysis Calculator. TensorFlow has allowed me to build and train robust models for large-scale machine learning projects.
I have used SQLite to store information for a deployed Django site. SQLite is a lightweight disk-based database that doesn't require a separate server process, making it an excellent option for storing data in applications, including web apps like Django. On the other hand, MongoDB is a NoSQL database that provides a scalable and flexible approach to data storage.
(Click image for more information)
A Command Line Interpreter
The Ship-Shell project is a hands-on exploration of the Linux shell environment and low-level system calls. The project is built around a new Linux shell dubbed "Ship Shell," which supports several built-in commands such as "cd", "path", and "exit". It also allows users to access commands from the "/bin" folder, change Path variables, and supports features such as parallelism "&" and redirection ">".
View this project on GitHub: https://github.com/Akeon201/ship-shell
Algorithm Analysis and Design
In this project, we explore the performance of two matrix multiplication algorithms, Strassen's and Brute Force, along with a hybrid approach that combines the strengths of both. Strassen's algorithm is a recursive method with a time complexity of O(n^2.807...), while the Brute Force algorithm has a time complexity of O(n^3). Recursion, as employed by Strassen's algorithm, often incurs considerable overhead, which can lead to significant time differences when processing smaller inputs. In some cases, Strassen's algorithm can take over ten times as long to compute matrix multiplication compared to the Brute Force approach.
The primary objective of this project was to identify an optimal cutoff value for the combination of Strassen's and Brute Force algorithms, ultimately maximizing the time saved during matrix multiplication. By carefully analyzing the performance of both algorithms under various input sizes, the project aimed to find a sweet spot where switching between the two methods would result in the most efficient and effective solution for matrix multiplication tasks.
In conclusion, the analysis of the combined algorithms for matrix multiplication demonstrates a steady decrease in the execution time as the cutoff value approaches 5. At this specific point, we observe the greatest time-saving performance. However, as the cutoff value increases beyond 5, there is a noticeable increase in execution time for values 6 through 8, indicating a diminishing return on efficiency. This trend highlights the point where the computational time required for the brute force algorithm begins to outweigh the benefits of Strassen's algorithm. Therefore, the optimal cutoff value of 5 is crucial for maximizing the efficiency and minimizing the execution time of the combined algorithm.
View this project on GitHub: https://github.com/Akeon201/Strassen-Algorithm-Research