Skip to content
View guyjuniorcalvet's full-sized avatar
  • Universite du Quebec a Chicoutimi
  • G7H 0S5

Block or report guyjuniorcalvet

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
guyjuniorcalvet/README.md

Hi there, I'm Guy Junior Calvet 👋

3rd-year Computer Science student · Data Science & Business Intelligence · UQAC

LinkedInEmailGitHubPortfolio


🧑‍💻 About Me

I'm a passionate Computer Science student at Université du Québec à Chicoutimi (UQAC), specializing in Data Science and Business Intelligence. I love building modern software solutions — from interactive web applications and data visualizations to machine learning models and database systems.

  • 🎓 B.Sc. Computer Science – Data Science & BI, UQAC (2023–2026)
  • 🎓 B.A. Business Administration, Université Notre Dame d'Haïti (2017–2021)
  • 🔍 Currently looking for an internship in software development, data science, or web development
  • 🌍 Based in Saguenay, QC, Canada — open to remote opportunities
  • 🤝 Member of Rotaract Club Delmas since 2020
  • 🏆 Participant at CS Games 2025 (ULaval)

🛠️ Tech Stack

Languages

PythonRC++JavaScriptTypeScriptSQLHTMLCSS

Frameworks & Libraries

FlaskStreamlitAstroTailwind CSSShinyTensorFlowKerasScikit--LearnNode.jsExpress

Databases & Cloud

MySQLSQLiteOracleMongoDBSupabaseGoogle Cloud

Tools & Practices

GitGitHub ActionsVS CodeJupyterAgile / Scrum


🚀 Featured Projects

Full-stack e-commerce web application built with Flask, Peewee ORM, and SQLite for the INF349 Advanced Web Technologies course. Features a REST API, product catalog UI, tax engine, shipping calculator, and online payment integration. Includes CI/CD with GitHub Actions and comprehensive pytest test suite.

PythonFlaskPeeweeSQLiteJinja2GitHub Actionspytest

Interactive R Shiny application for data entry and management of the Rotaract de Delmas database, connected to a cloud-hosted MySQL database. Built during a university internship.

RShinyMySQLGoogle CloudSupabase

Two machine learning classification projects: a Multilayer Perceptron (MLP) with Scikit-Learn for tabular data and Convolutional Neural Networks (CNN) with TensorFlow/Keras (including VGG16 transfer learning) for image classification.

PythonTensorFlowKerasScikit-LearnCNNVGG16Jupyter Notebook

Data visualization and interactive interface project analyzing Spotify data using R.

RData Visualization

Personal portfolio web app built with Streamlit, featuring bilingual support (FR/EN), project showcase, CV download, and an AI-powered chatbot that generates tailored cover letters using OpenAI GPT.

PythonStreamlitOpenAI API

Modern portfolio website rebuilt with Astro and Tailwind CSS, deployed on GitHub Pages. Uses TypeScript for configuration and features a clean, fast, static site architecture.

AstroTailwind CSSTypeScriptGitHub Pages

Implementation of postfix notation evaluation using stacks and data structures in C++.

C++Data Structures

Implementation of a professor file management system using linked lists and data structures in C++.

C++Data Structures

Interactive 3D visualization of Keycube study data, deployed as a static web application on GitHub Pages. Exposes two autonomous views — Preference (participant-by-participant preference data + aggregate) and Reachability (per-finger and combined heatmaps) — rendered directly on a 3D model of the device. CSV datasets are transformed into browser-usable data at build time via a Python script (build_local_site.py). Supports English and French page variants. No Node.js or external Python packages required.

JavaScriptThree.jsHTMLPythonCSVWebGLGitHub Pages

Team project (Oscar Neveux · Wilson Fotsing · Guy Junior Calvet · Jean-Christophe Barriault) for the Data Mining course at UQAC.

Automated classification of cybersecurity incidents (TruePositive vs Non-TruePositive) using the Microsoft GUIDE dataset (~2.4 GB), designed to help SOC analysts prioritize alerts at scale. The pipeline handles stratified sampling, NaN imputation, cardinality reduction (MITRE ATT&CK techniques, top-N grouping), temporal feature engineering, and full preprocessing exports. Three models were trained and compared — Logistic Regression + PCA, KNN + PCA, and Random Forest — with the Random Forest selected for production thanks to its best recall (fewest missed incidents). A Streamlit interface enables CSV-based batch analysis, manual form input, and KPI / ROC / PR visualizations.

PythonScikit-LearnRandom ForestLogistic RegressionKNNPCAStreamlitJupyter NotebookMITRE ATT&CK


📈 GitHub Stats

GitHub StatsTop Languages


📫 Get in Touch


💡 I see challenges as opportunities to grow — one step at a time.

Profile views

Pinned Loading

  1. Rotaract_delmas_ShinyRotaract_delmas_ShinyPublic

    Application d'alimentation (Shiny) et de gestion de la base de données du Rotaract de Delmas

    R 1

  2. Spotify_Visualisation_et_interfaceSpotify_Visualisation_et_interfacePublic

    R

  3. Reseaux_de_neurones_LMReseaux_de_neurones_LMPublic

    Deux projets de classification avec des réseaux de neurones : un PMC avec Scikit-Learn pour des données tabulaires et des CNN avec TensorFlow/Keras pour des images.

    Jupyter Notebook

  4. Mon_PortfolioMon_PortfolioPublic

    Mon portfolio avec Streamlit

    Python 1

  5. keycube/heatmap3dkeycube/heatmap3dPublic

    The web site adress

    HTML