Economist and Data Scientist

It's amazing how we can currently connect with a large number of people around the world. Today you are here, getting to know a part of my passion and work.

- Data Science empowers us to unlock valuable insights from complex data, driving innovation and informed decisions. It transforms raw information into actionable knowledge, enabling success and growth in every project

About me

I hold a degree in Economics from Universidad del Valle, Colombia, with a specialization in Data Analysis and Science. My passion for data analysis and artificial intelligence has driven me to immerse myself in Machine Learning (ML) and work on innovative projects. My academic background enables me to blend economic theory with advanced data analysis techniques and predictive modeling tools, including supervised and unsupervised learning algorithms, regression models, classification, clustering, and deep learning. I am proficient in Python, R, and libraries such as Scikit-learn, TensorFlow, and Keras.

Throughout my career, I have developed projects involving large dataset analysis, predictive modeling, and process automation. My interests extend to natural language processing (NLP), neural networks, and time series analysis, which has deepened my understanding of how data can transform decision-making across sectors. I aim to continue advancing my expertise in Machine Learning and Big Data, exploring emerging technologies, and applying innovative solutions to solve complex, real-world challenges.

Skills

SQL for Data Extraction & Analysis
I develop optimized queries in MySQL and SQL Server, including complex joins, aggregations, and data cleaning for reports and dashboards.
Interactive Dashboards with Power BI
I create dynamic, business-oriented dashboards: data modeling, DAX calculations, and interactive visualizations to support decision-making.
R for Statistical Modeling & Visualization
I apply statistical techniques, conduct hypothesis testing, and craft advanced visualizations with ggplot2 and the tidyverse.
Python for Data Science & Machine Learning
I build machine learning pipelines: preprocessing, exploratory analysis, and model evaluation using scikit-learn, pandas, and matplotlib.
Automation with Excel Macros & Python
I develop VBA tools and Python scripts to automate reporting, file processing, and SharePoint audits.

My projects

PYTHON

This project focuses on the analysis and supervised classification of the well-known Fisher's Iris dataset, aiming to accurately predict the species of iris flowers (Iris setosa, Iris versicolor, and Iris virginica) based on numerical features: sepal length, sepal width, petal length, and petal width.

This web application recommends TEDx talks similar to a text provided by the user.
It leverages TF-IDF and cosine similarity (along with Pearson correlation) to analyze the content and generate a list of the most relevant talks, presenting the user with each talk's main speaker and description.

This web application predicts possible diseases based on user-selected symptoms.

It leverages machine learning algorithms — SVM, Naive Bayes, and Random Forest — to analyze the input and generate the most probable diagnosis, providing consistent and reliable results in real time.

Automated consolidation and rule-based auditing of commercial & marketing data (sales, leads, campaigns, pricing, discounts), using Excel-defined business rules (price floors, discount caps, bonuses) to normalize disparate files, flag anomalies (zero-conversion spend, out-of-policy discounts, negatives, gaps), and deliver filterable reports with one-click exports.

Real-time author identification from scraped quotes, combining text normalization and typo-tolerant fuzzy matching (SequenceMatcher ≥ 0.85), with on-demand hints (birth date/place) and Wikipedia-based image enrichment (graceful fallbacks), exposed via a lightweight Flask API with in-memory caching.

Automated QA for campaigns: verifies required assets in folder structures, checks month/year headers, validates parameterized ratios against expected values, reconciles line items with typo-tolerant similarity, and applies temporal rules, then outputs sectioned, audit-ready reports.

Automated discovery, consolidation, and rule-based auditing of distributed enterprise spreadsheets (plans, trackers, catalogs, costs), using template-defined checks (sheet presence, naming drift, key normalization, code reconciliation) to standardize inputs, flag anomalies (missing tabs, mismatched codes, out-of-range totals, broken formulas), and produce executive-ready workbooks with preserved formatting, role-based visibility, versioning, multi-location publishing, and API-triggered one-click runs.

AUTOMATED SOLUTIONS FOR CLIENTS

An intuitive Excel template that automates your invoicing and quoting: it captures key data, converts totals into Spanish text with currency codes, auto‑generates three‑digit IDs, and exports print‑ready PDFs with smart, data‑driven file names.

This file is an automated Excel template designed to manage and document technical projects related to installation, maintenance, or infrastructure expansion, especially in the telecommunications sector. It is intended for internal use by companies that carry out technical site visits, install equipment, and track detailed information on resources, materials, and project progress by client or by city.

This project focuses on the analysis and supervised classification of the well-known Fisher's Iris dataset, aiming to accurately predict the species of iris flowers (Iris setosa, Iris versicolor, and Iris virginica) based on numerical features: sepal length, sepal width, petal length, and petal width.