About me

I am a Data Scientist at the EdTech Node of Universidad de los Andes, where I am also pursuing an M.Sc. in Information Engineering. My research focuses on making language models reliable for the people and languages they serve worst: multilingual NLP and cross-lingual transfer for low-resource languages, and how to evaluate models where labeled data is scarce, from label-free model selection to LLM-as-judge, rubric-based evaluation, and culturally grounded benchmarks. I pursue this line as a collaborating researcher at the Lee Language Lab (Ontario Tech University), which I joined through a Mitacs Globalink Research Internship.

In parallel, I build production data and machine learning systems for higher education: learning analytics, student-success prediction, and university-wide data products that support academic advising, work that has led to peer-reviewed publications at AIED and WEEF. More broadly, I care about human-centered AI that serves people, from underrepresented language communities to university classrooms.

In my free time, I enjoy playing chess and bowling.