I'm a Data Scientist & LLM Engineer based in Monterrey, México, currently finishing my M.Sc. in Data Science at CIMAT. My work lives at the intersection of Natural Language Processing, LLM systems, and MLOps. Right now I'm leading research on tourist review analysis for Mexican destinations, building RAG pipelines, and thinking a lot about how to make language models actually useful in the real world.

RAG system that turns scattered tourist reviews into strategic intelligence reports for México's Pueblos Mágicos.

Built a full LangGraph Map-Reduce workflow that retrieves reviews from a ChromaDB vector store, extracts structured insights per business type in parallel, and consolidates them into a validated strategic briefing — with a self-correction audit loop and an interactive Streamlit + FastAPI interface on top.

End-to-end MLOps pipeline for loan eligibility prediction, deployed on AWS.

Full ML lifecycle: experiment tracking with MLflow, pipeline orchestration with Prefect, cloud infrastructure provisioned with Terraform, and real-time serverless inference via AWS Lambda + Kinesis. Includes unit tests, LocalStack integration tests, and automated linting... the kind of project where the engineering matters as much as the model.

Hybrid Two-Tower architecture that tackles the cold-start problem in music recommendation.

Combined content-based multimodal encoders with sequential user modeling to recommend music even for new users or tracks with no history.

At REST-MEX 2025, a national NLP competition focused on Mexican tourism reviews, our team placed 7th out of all participating teams, earning an honorable mention. Good times.

I'm always open to interesting conversations about NLP, LLM systems, or MLOps. Feel free to reach out!