frank photo

Hi, I'm Francisco

Computer Engineer and Data Scientist with experience in full-stack development and data modeling. Focused on clear, efficient, and evidence-based solutions.

Education

  1. •

    M.Sc. in Data Science and Information

    INFOTEC, Center for Research and Innovation in Information and Communication Technologies
    August 2023 — December 2025
  2. •

    B.Sc. in Computer Science Engineering

    Benemérita Universidad Autónoma de Puebla
    August 2015 — December 2020

Experience

  1. •

    Full-Stack Developer

    MTech Systems

    Contributed to the development of a professional poultry management web platform, working on both frontend and backend tasks within an agile team. Supported the implementation and improvement of business-oriented features. Coordinated technical work in a bilingual environment, enabling effective communication among team members. Learn more...
  2. •

    Full-Stack Developer

    beedSocial

    Served as a Full-Stack Developer leading the development of a web application. Participated in database design and structuring, as well as in the implementation of the system API to manage business logic and communication. Collaborated closely with the development team to incorporate improvements and new features aligned with product evolution.
  3. •

    Front-End Developer

    beedSocial

    Contributed to the development of a web application for career guidance, participating in the design and implementation of the user interface. Adapted components and functionalities as product requirements evolved, ensuring consistency and usability within a collaborative development environment.
  4. •

    IT Support

    SAT - Puebla 1 Decentralized Administration (Tax Administration Service)

    Provided IT support at the Puebla 1 Decentralized Administration, participating in IP address mapping of institutional computing equipment and verification of switch ports to ensure proper network operation. Performed software installations and updates, as well as direct technical support to users for resolving operational issues.

Projects

Interface for ΔG Genetic Data Analysis

Developed an interactive R Shiny interface for the analysis of genetic sequences with energy values (ΔG), enabling sequence grouping by similarity, application of both automatic and manual clustering methods, and exploration of results through interactive visualizations such as dendrograms. Additionally, a tool was implemented to generate sequence combinations for structural biology analysis. The project aims to facilitate access to advanced genomic analysis tools for researchers and healthcare professionals without requiring advanced programming skills.

  • R language
  • Shiny

Platform for Evaluation and Interpretability of Machine Learning Models in Infant Neurodevelopment

Developed an interactive interface for the comparative evaluation of supervised models applied to infant neurodevelopment prediction, as a complement to a scientific outreach article on clinical Machine Learning. The tool enables analysis and comparison of models such as Logistic Regression, Decision Trees, Random Forest, and XGBoost using metrics including accuracy and F1-score. The interface integrates key visualizations for model validation, including confusion matrices, AU-ROC curves, and explanations based on SHapley Additive exPlanations (SHAP), facilitating interpretation of both performance and predictor contributions. Additionally, it incorporates configurable options for hyperparameter tuning and oversampling techniques, which can be enabled independently or in combination during the training process. The project is framed within the prediction of Bayley scores (PDI and MDI) from clinical, perinatal, and anthropometric data, supporting early identification of risk factors and informed decision-making in real-world clinical settings.

  • R language
  • Shiny

Articles

  1. •

    Predictive Model of Infant Neurodevelopment with Katona Therapy: PDI and MDI Prediction Using Machine Learning

    December 2025

    Machine Learning applied to the prediction of infant neurodevelopment. Development of supervised models (Random Forest and XGBoost) to predict Bayley scores (PDI and MDI) from clinical, perinatal, and anthropometric data. The project involved data cleaning and preprocessing, imbalance handling, feature selection, and hyperparameter optimization, identifying early variables with high predictive power. Results demonstrate how Machine Learning can support risk stratification and decision-making in real-world clinical contexts.

About me

I am a Computer Science Engineer and a recent graduate of the M.Sc. in Data Science and Information at INFOTEC, a public research center of the Mexican Federal Government. My background combines software development with data analysis, allowing me to approach problems from a technical and structured perspective.

I have worked as a full-stack developer in both enterprise and product-oriented environments, contributing to the design and implementation of web applications, databases, and APIs. This experience provided me with a practical understanding of real-world systems: how they are built, how they evolve, and how technical decisions impact users and organizations.

My current focus is on Machine Learning and data analysis, with interest in both practical applications and research. I am motivated to build data-driven solutions that are technically robust, interpretable, and useful in real-world contexts, especially when problems demand methodological rigor and critical thinking.

Francisco Ferrusca