Programming & Analysis
PythonSQLPandasNumPyStatistical AnalysisData Cleaning
Data Analyst · Decision Scientist
Turning complex data intoactionable business insights.
I'm Matias Torres — a Data Analyst and Decision Scientist who transforms raw data into clear, strategic insights using SQL, Python, Power BI, Tableau, statistics, and machine learning.
ETL
Cleaning & Transformation
Analytics
Statistics & ML
Visualization
Tableau & Power BI
Data quality & QA
ML workflows
AI-assisted analytics
APIs & MCP
About
Connecting data, technology, and business decisions.
I'm a Data Analyst and Decision Scientist with a multidisciplinary background in Data Science, Business Analytics, and Business Management. I use SQL, Python, Power BI, and Tableau to turn raw information into insights that support strategic decisions.
In AI data and data science projects, I've validated datasets, reviewed model outputs, assured data quality, and evaluated AI-generated Python workflows. My work has included supervised learning, business dashboards, Generative AI tools, REST APIs, and MCP servers.
I also bring hands-on business experience from managing sales, inventory, suppliers, marketing, and performance analysis at Dos Torres Sport, alongside freelance analytical projects. That combination helps me communicate clearly with technical and non-technical stakeholders and keep solutions grounded in real business needs.
Business Analytics
ML & Statistics
Generative AI
Business Operations
Skills
The tools I use to turn data into decisions.
A practical stack spanning analysis, visualization, machine learning, and AI-assisted workflows.
Machine Learning
Scikit-learnClassificationRegressionModel EvaluationFeature Engineering
Data Visualization
TableauPower BIBusiness DashboardsInsight Communication
Artificial Intelligence
Prompt EngineeringGenerative AILangGraphLangChainREST APIsMCP Servers
Case Studies
Projects with measurable business impact.
Every project is framed as a case study: the problem, the approach, and the outcome.
Featured case study · SQL & Tableau
Customer RFM Segmentation & Marketing Dashboard
An end-to-end customer analytics project that transforms raw campaign data into an RFM segmentation model, dashboard-ready dataset, interactive Tableau analysis, and targeted marketing recommendations.
2,201
customers analyzed
78.53%
value in 3 priority segments
14.99%
last campaign acceptance
Loyal, At Risk, and Can't Lose Them customers account for 78.53% of historical customer value, making retention and reactivation the clearest business opportunity.
SQLPostgreSQLDockerRFMTableau
Featured case study · Python & Machine Learning
Customer Churn Prediction
An end-to-end supervised learning analysis of Online Retail II transactions. After cleaning duplicates, unidentified customers, cancellations, and non-product activity, I engineered RFM, recent-activity, tenure, geography, and cancellation features from a 365-day lookback to predict 180-day purchase inactivity. Logistic regression, random forest, and histogram gradient boosting were compared with five-fold cross-validation.
4,324
customers analyzed
0.791
holdout ROC-AUC
86.4%
churner recall
A training-only F1 threshold of 0.352 prioritized recall: the final model identified 325 of 376 churners on the untouched holdout set, supporting a recall-oriented retention strategy.
PythonPandasScikit-learnJupyteruv
Resume
Experience & background.
Full CV — PDF
Detailed experience, education, and selected projects.
AI Data Annotator & Data Science Contributor
Remote projects · 2024 — 2026
Co-Owner & Business Manager
Dos Torres Sport · Jan 2020 — Present
Freelance Data Analyst
Independent · Jan 2021 — Jan 2023
Data Science & Data Analytics
CoderHouse · 2023
Contact
Have a problem worth solving?
I'm happy to discuss projects, collaborations or full-time roles.