Skip to content
View DiegoXavier-hub's full-sized avatar
:octocat:
:octocat:

Block or report DiegoXavier-hub

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
DiegoXavier-hub/README.md

Diego Xavier

Typing SVG

Data Science student from Brazil focused on turning data into analysis, models, and decision-ready insights.

About Me

  • Focused on data analysis, machine learning, and reproducible workflows
  • Interested in predictive modeling, exploratory analysis, interpretability, and data visualization
  • I enjoy building projects that connect data, code, and clear technical communication
  • Also comfortable shipping full-stack web apps to support and showcase analytical work
  • Always learning and improving through hands-on analytical work

Current Interests

Predictive Modeling Survival Analysis EDA Feature Engineering Model Evaluation Interpretability Hyperparameter Optimization Data Visualization Analytical Pipelines Full-Stack Web Technical Writing

Tech Stack

Core Languages

Python TypeScript JavaScript HTML5 CSS3 Jupyter LaTeX

Data Science and Machine Learning

Pandas NumPy Scikit-Learn XGBoost LightGBM CatBoost Optuna SHAP Lifelines SciPy Statsmodels PyArrow OpenPyXL

Visualization

Plotly Matplotlib Seaborn Excel

Web - Frontend

Next.js React Vite React Router Lucide

Web - Backend

Node.js Express Firebase Firestore JWT Zod Nodemailer

Testing and Tooling

Vitest Testing Library ESLint Git GitHub Markdown

Featured Projects

  • TCC - Modelagem Preditiva — Pipeline completo com CatBoost, LightGBM e Logistic Regression, otimização com Optuna, calibração de threshold, interpretabilidade via SHAP e análise de robustez.
  • Radar de Risco - Sobrevivência de CNPJs — Análise de sobrevivência de empresas brasileiras (Kaplan-Meier, Cox) combinada com modelagem de risco de baixa em 12 meses e storytelling analítico.
  • ML from Scratch — Implementações em Python puro/NumPy de Decision Tree, Random Forest, KNN, KMeans, Linear/Logistic Regression, Naive Bayes, PCA, Perceptron e SVM.
  • Portfólio Pessoal (Full-Stack) — Frontend em Next.js 16 + React 19 + TypeScript, backend em Express 5 + Firebase Admin + Firestore, com autenticação JWT, validação Zod e testes com Vitest.
  • Dos Santos Plumbing Services — Site institucional em Vite + React 19 com React Router, otimização de imagens via Sharp e pipeline de testes com Vitest + Testing Library.

GitHub Overview

Contribution graph
Top languages GitHub streak

What I Like Building

  • Analytical projects with strong documentation and reproducibility
  • Predictive models with practical evaluation and interpretation
  • Clear visual narratives that make data easier to understand
  • Full-stack web apps that turn analyses into usable products
  • Workflows that go from raw data to final report with consistency

Always learning, building, and improving through data.

Pinned Loading

  1. LLMs-sob-o-microscopio LLMs-sob-o-microscopio Public

    Python

  2. Sobrevivencia-de-empresas-brasileiras Sobrevivencia-de-empresas-brasileiras Public

    Python

  3. Machine-Learning-from-scratch Machine-Learning-from-scratch Public

    Jupyter Notebook