Carlos Alfredo Torres Cubilla

Carlos Alfredo Torres Cubilla

Data Scientist

Biography

Carlos Torres is a Data Scientist with a strong background in Statistics, Machine Learning, and Big Data Analytics.

He holds both a Bachelor’s degree in Statistics and a Master’s degree in Advanced Analysis of Multivariate Data and Big Data from the University of Salamanca, where he developed solid skills in programming, predictive modeling, and applied research.

Carlos has experience working with cloud platforms such as AWS and Databricks, where he has trained and deployed machine learning models for real-world applications. Previously, he worked at Banco General, where he developed data-driven solutions to optimize loan recovery and increase the bank’s profitability.

He is currently focused on expanding his expertise in advanced analytics and data-driven decision-making to develop impactful and efficient solutions, while continuing to explore new applications of artificial intelligence and machine learning in real-world contexts.

Download my CV .

Interests
  • Statistics
  • Artificial intelligence
  • Machine learning
  • Programming
Education
  • Master's Degree in Advanced Analysis of Multivariate Data and Big Data, 2020

    University of Salamanca

  • Bachelor's degree in Statistics, 2019

    University of Salamanca

Technical skills

Python

Development of machine learning models, task automation, and data analysis

R

Advanced statistical analysis, visualization, and predictive modeling

databricks
Databricks

Building and deploying data and machine learning pipelines in collaborative environments

mlflow
MLflow

Experiment management and production deployment of machine learning models

AWS

Experience with S3, EC2, and QuickSight for cloud-based analytics

Git

Version control, collaborative workflows, and reproducible development

SQL

Basic understanding to interpret and work with queries

Markdown

Clear and structured technical documentation in reproducible environments

HTML

Basic knowledge of HTML structure for presenting results or integrating content

Power skills

Analytical Thinking

Ability to analyze information and make data-driven decisions

Problem Solving

Critical and creative approach to addressing complex challenges

Adaptability

Flexibility in dynamic environments and openness to new technologies

Continuous Learning

Commitment to lifelong learning and professional growth

Effective Communication

Clarity and empathy when conveying technical and strategic ideas

Leadership Potential

Ability to motivate, guide, and coordinate teams toward common goals

Experience

 
 
 
 
 
Banco General
Panama City, Panama
Jan 2021 – Feb 2025
Senior Data Scientist

Apr 2022 – Feb 2025

  • Led strategic advanced analytics projects in banking, such as Delinquency Forecasting and Recovery Optimization, achieving substantial improvements in portfolio recovery and delinquency reduction.
  • Developed and implemented machine learning models (Scikitlearn, LightGBM, XGBoost) deployed in production with scalable pipelines in Databricks and monitoring through MLflow.
  • Designed robust ETL pipelines in PySpark to feed executive dashboards in AWS QuickSight, optimizing data-driven decision-making for senior managers.
  • Mentored a team of data scientists and engineers, promoting data engineering and visualization standards aligned with international best practices.
Data Scientist

Jan 2021 – Mar 2022

  • Participated in end-to-end analytics projects such as predicting credit card churn, insurance sales, and next best product to buy, integrating predictive models directly into business processes for collections and risk.
  • Automated ETL processes in PySpark and developed predictive models in Python, increasing the scalability and efficiency of analytical flows.
  • Prepared financial reports with visualizations tailored to technical and non-technical audiences, facilitating the communication of findings and recommendations.
 
 
 
 
 
STAT-UP
Munich, Germany
Data Science Intern
Feb 2020 – Apr 2020
  • Implemented ETL processes in R (Tidyverse) to consolidate and clean data from various sources.
  • Developed and deployed an interactive application in Shiny for multivariate statistical analysis, improving data exploration and visualization.
  • Created interactive HTML reports with dynamic visualizations, optimizing the communication of results to international clients.
 
 
 
 
 
Statistics Intern
Sep 2018 – Jan 2019
  • Advised students on the application of advanced statistical techniques to academic projects.
  • Supported teaching in applied statistics courses, promoting the use of analytical tools and statistical software.

Projects

*
SparseBiplots

SparseBiplots

R package that performs the HJ-Biplot and modifications introducing Ridge, LASSO and Elastic Net penalty.

Covid-19 in Castilla & Leon

Covid-19 in Castilla & Leon

Web app to visualize the epidemiological status of COVID-19 in Castilla & León, Spain.

PyBiplots

PyBiplots

Python library that performs the classic biplots methods.

HackForGood 2019

HackForGood 2019

Project carried out to solve the challenge of Artificial Intelligence for the improvement of the Health System.

Recent Publications

Quickly discover relevant content by filtering publications.
(2022). What environmental sustainability practices do universities manage for sustainable development. 2022 8th International Engineering, Sciences and Technology Conference (IESTEC).

Cite DOI

(2021). Sparse HJ Biplot: A New Methodology via Elastic Net. Mathematics, 9(11).

Cite Project Journal DOI PDF

Contact