Research

Doctoral focus

I'm a first-year PhD student in the joint Artificial Intelligence doctoral program of the CRUCH Biobío-Ñuble consortium (UdeC, UBB, UTFSM, UCSC), with an internal scholarship. My research sits at the intersection of AI and education.

The question driving my work: as large language models become embedded in how students learn, what patterns of student–AI interaction actually emerge in the wild, and what do they imply for learning outcomes and equity?

Research interests

• Student–AI interaction in higher education. Characterizing how undergraduate engineering students use LLMs in coursework — when they delegate vs. dialogue, what they ask, where they get stuck, how their use evolves across a semester.
• Causal effects on learning outcomes. Designing classroom-grounded experiments and quasi-experiments to move past "students who use AI do X" toward defensible claims about impact.
• Educational equity. Examining whether AI tools amplify or compress prior gaps: by prior preparation, language, gender, socioeconomic background.
• Generative AI in engineering education. Curriculum and assessment design that takes seriously what's now easy and what's now harder to evaluate.

Prior research

Before the PhD, my research used social media data to study gender differences in how people perceive and talk about urban transportation. That work appeared in Sustainability (2020) and at the ACM Web Science Conference (WebSci '19), and led to invited talks at the Inter-American Development Bank's Transport GenderLab.

Methodologically that line taught me a lot about working with observational text data, building careful operationalizations of social constructs, and translating findings for policy audiences — all of which carries over to the AI-in-education work.

Methods I work with

• Experimental design and causal inference
• Mixed-methods analysis of student interaction logs
• NLP for educational discourse
• Multilevel regression with post-stratification (from my Telefónica R&D internship)