Skip to content
View MapiAI's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report MapiAI

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
MapiAI/README.md

👋 Hi, I’m Maria Petralia (MaPi)

About Me

I'm a Data Scientist and AI practitioner with a background in Computer Science and more than ten years of experience in software development and digital solutions. I build end-to-end data and AI projects spanning machine learning, predictive analytics, natural language processing, computer vision, and generative AI applications.

My portfolio includes customer segmentation, time series forecasting, NLP classification, recommendation systems, and computer vision projects, with a strong focus on feature engineering, model evaluation, explainability, and reproducible workflows. I enjoy transforming complex data into actionable insights and designing AI-powered solutions that combine technical rigor with real-world impact.

I am particularly interested in Machine Learning, Deep Learning, Natural Language Processing, Computer Vision, Explainable AI, Large Language Models, Retrieval-Augmented Generation (RAG), and the development of responsible AI systems that address meaningful business and social challenges.


Interests

  • Machine Learning & Artificial Intelligence
  • Computer Vision & Deep Learning
  • Natural Language Processing & Text Classification
  • Large Language Models & Generative AI
  • Retrieval-Augmented Generation (RAG)
  • Responsible AI & Ethical AI Design
  • Data Analytics & Visualization
  • Predictive Modeling & Time Series
  • Unsupervised Learning & Clustering
  • Explainable AI & Model Evaluation
  • Reproducible Data Science Workflows

Featured Projects

TravelTide – Customer Segmentation

Unsupervised learning project using segmentation and clustering to identify customer personas and behavioral patterns.
🔗 Repo: https://github.com/MapiAI/TravelTide-Customer-Segmentation

Corporación Favorita – Time Series Sales Forecasting

End‑to‑end forecasting project with MLflow tracking, Streamlit apps, and model evaluation.
🔗 Repo: https://github.com/MapiAI/corporacion_favorita

CIFAR-10 Image Classification (CNN, ResNet50 Transfer Learning)

Image classification on the CIFAR-10 dataset using a custom CNN and a transfer learning approach based on a pretrained ResNet50 model.
🔗 Repo: https://github.com/MapiAI/cifar10-image-classification

Vehicle Silhouette Classification

Supervised machine learning project that classifies vehicle types using tabular features extracted from vehicle silhouettes.
🔗 Repo: https://github.com/MapiAI/Vehicle-Silhouette-Classification-Project

Car Data Analysis (Python, Jupyter)

Exploratory data analysis and visualization on car datasets.
🔗 Repo: https://github.com/MapiAI/Car-Data-Analysis

Tableau Dashboards


Currently Learning

  • Generative AI & Conversational AI Systems
  • Large Language Models (LLMs) and Prompt Engineering
  • Deep Learning & Computer Vision
  • AI Application Development
  • Model Deployment and MLOps Foundations
  • Reproducible AI and Machine Learning Workflows

Collaboration

I’m open to collaborating on data science projects that combine technical depth with clear communication and visualization.


How to Reach Me


Fun Fact

I find beauty in both nature and numbers. When I’m not analyzing data or creating vector images, you’ll find me designing cakes 🎂, tending to my garden 🌱, or practicing yoga & tai chi 🧘‍♀️. Creativity and logic are the two forces that shape my world.

Pinned Loading

  1. TravelTide-Customer-Segmentation TravelTide-Customer-Segmentation Public

    Customer segmentation and perk assignment project for TravelTide, combining rule-based behavioral logic and clustering.

    Jupyter Notebook

  2. DisasterTweets-NLP-Classification DisasterTweets-NLP-Classification Public

    NLP binary classification pipeline to detect disaster-related tweets - feature engineering, TF-IDF vectorization, and linear models.

    Jupyter Notebook

  3. Vehicle-Silhouette-Classification-Project Vehicle-Silhouette-Classification-Project Public

    Supervised machine learning project for classifying vehicles (car, van, bus) using geometric features extracted from their silhouettes. Includes EDA, preprocessing, multiple model evaluations, cros…

    Jupyter Notebook

  4. corporacion_favorita corporacion_favorita Public

    Time series sales forecasting for Corporación Favorita — XGBoost, SARIMA, LSTM, Streamlit app

    Jupyter Notebook

  5. Car-Data-Analysis Car-Data-Analysis Public

    Car dataset analysis using Python, Pandas, and visualization libraries. Exploratory Data Analysis (EDA) uncovering insights on pricing, performance, efficiency, and automotive market trends.

    Jupyter Notebook 2

  6. cifar10-image-classification cifar10-image-classification Public

    Image classification on CIFAR-10 using a baseline CNN and ResNet50 transfer learning with fine-tuning, augmentation, and error analysis.

    Jupyter Notebook