Data tells stories... I translate them into human decisions that transform organizations.

Language Teaching + Data Science = Tech for Good with Impact

I combine my background in communication and language teaching with advanced data science expertise to build analytical and predictive solutions that optimize environmental sustainability, healthcare workflows, and social impact.

My differentiator: I understand both the complex statistical models and the human communication needed to drive global change.

View Projects Download CV

Fiorella Trigo

Email fiorellatrigo@gmail.com

LinkedIn linkedin.com/in/fiorella-trigo

GitHub github.com/fiorellatrigo

Languages Spanish (Native) | English (C2) | Italian (B1) | Portuguese (B1)

Location Buenos Aires, Argentina

Contact me

Projects

Wind Energy Prediction

Python Pandas XGBoost Scikit-Learn Time-Series Data Restructuring Chronological Validation Split Model Evaluation (MAE, RMSE, R²) Matplotlib / Seaborn

A data study to forecast Germany's wind energy output. I transformed complex daily spreadsheets into a continuous timeline and trained an XGBoost model to find generation patterns. By reducing prediction errors by 37 MW, this project helps grid operators and traders prevent energy waste and balance clean energy distribution.

GitHub Repo

Video Game Industry Analysis: Strategic Insights for 2017

Python Pandas Seaborn Matplotlib Exploratory Data Analysis Statistical Testing

This project performs a comprehensive data analysis for the online store Ice, which sells video games globally. The primary objective is to identify patterns that determine whether a game will be successful or not, enabling the store to plan its advertising budget and strategic inventory for the upcoming year (2017).

GitHub Repo

Chicago Taxi Market & Weather Impact (Nov. 2017)

Python SQL Scipy (Stats) Hypothesis Testing Matplotlib

A comprehensive study of Chicago's transport dynamics. I analyzed market share among taxi companies, identified geographic hotspots like the "Loop", and used Independent T-testing to prove how adverse weather significantly increases trip duration to O'Hare Airport.

GitHub Repo

Megaline Consumer Revenue and Usage Analysis

Python Pandas NumPy Exploratory Data Analysis (EDA) Statistical Hypothesis Testing Scipy (t-test, Levene) Matplotlib

A data-driven analysis of Megaline’s prepaid plans. I examined usage patterns across calls, messages, and mobile data, and applied statistical hypothesis testing to demonstrate how overconsumption — particularly of internet data — drives higher average revenue in lower-priced plans.

GitHub Repo

Megaline Subscriber Classification Model

Mobile Plan Subscriber Classification Model

Python Pandas Scikit-Learn Supervised Learning Classification Algorithms Hyperparameter Tuning (GridSearchCV) Ensemble Methods (Random Forest, Gradient Boosting) Model Evaluation (Accuracy, Recall, F1) Matplotlib / Seaborn

A predictive modeling study for Megaline’s subscriber base. I implemented and tuned various classification architectures to automate plan recommendations. By capturing non-linear consumption patterns, I developed a high-accuracy engine (82%) that transitions the business from reactive auditing to proactive, AI-driven subscriber optimization.

GitHub Repo

Skills

Python

I use Python to clean, analyze, and explore data efficiently. I’m comfortable working with libraries like pandas and NumPy, writing clear scripts, and turning raw datasets into structured, usable information for analysis.

Data Visualization

I create clear and meaningful visualizations that help data tell a story. I use charts and dashboards to highlight patterns, trends, and insights, making complex information easy to understand for both technical and non-technical audiences.

SQL & Databases

I use SQL to query, filter, and aggregate data from relational databases. I’m confident writing joins, subqueries, and grouped analyses to extract exactly the data needed for exploration and decision-making.

Data Wrangling & Cleaning

I specialize in transforming complex, messy, and unstructured datasets into clean, reliable pipelines. I handle missing values, anomalies, and data alignment to ensure subsequent analytical and modeling phases are highly accurate.

Machine Learning

I build predictive models to uncover patterns and forecast trends. I apply feature engineering, strict validation, and algorithms like Random Forest or XGBoost using Scikit-Learn to deliver honest, data-driven solutions.

Data Storytelling

I focus on explaining insights in a clear and engaging way. I translate data results into narratives that answer real questions, support decisions, and connect numbers to real-world context.