ABOUT
I am a quantitative economist from Colombia with an M.A. in Economics and a Minor in Computational Mathematics from Universidad de los Andes. My work sits where economics, computation and social data meet: I build the pipelines and models needed to measure social phenomena, and I use causal inference to understand what policies actually do.
Research interests
Computational social science · migration and online discourse · causal inference for public policy · complex systems and networks · NLP for Spanish-language social data.
Highlights
- Documento CEDE No. 42 (2022). My master’s thesis, Safe Isolation: Ingreso Solidario and Bancarization, was selected by the School of Economics for its student working-paper series. SSRN
- Doctoral research on migration at Harvard (2022). I designed and ran the LDA topic-modelling analysis of mayoral candidates’ platforms along the Venezuelan migrant route for Olivia Woldemikael’s dissertation, where I am acknowledged for research assistance. Dissertation
- El Barómetro de Xenofobia (2020–2023). As Senior Analyst, I built social-media data pipelines and Spanish NLP models to measure online xenophobia. The project won 1st place in the inaugural Youth Leadership & Innovation Award for Migration at the 2021 Global Forum on Migration and Development. GFMD programme
- tidyX. Co-author of an open-source Python package for preprocessing Spanish social-media text. PyPI
Right now
- Data Engineer at Omnicom Global Solutions (since 2024): production pipelines and applications with Python, Dagster, dbt, Snowflake, React and AWS, including LLM-assisted feature engineering.
- Teaching Assistant at Universidad de los Andes (since 2022): analytics and programming courses in the Department of Industrial Engineering.
- Learning: computer-science coursework at Metropolia University of Applied Sciences (C, data structures and algorithms, discrete mathematics, operating systems, microprocessors, SQL, Unix/Linux) and German at the Goethe-Institut.
Toolbox
Research: causal inference (difference-in-differences, event studies, panel data), econometrics, topic models, NLP, statistical modelling. Code: Python, SQL, Java, JavaScript, Stata, LaTeX. Data and ML: Snowflake, Dagster, dbt, Spark, AWS, PostgreSQL, TensorFlow, spaCy, gensim, NLTK, LLM and RAG workflows. Software: React, Flask, Dash, Django, FastAPI, Docker, Git.
Languages
Spanish (native) · English (C1, TOEFL iBT 101) · German (B2 in progress)