From first principles
to production systems

I'm a Machine Learning Engineer based in Brasília, with 5+ years of experience evolving from data science toward ML engineering. At Centro de Gestão e Estudos Estratégicos (CGEE), I develop and deploy end-to-end data analytics applications using Python, FastAPI, and Docker, manage the Linux server infrastructure, and integrate LLMs into backend services through RAG pipelines and fine-tuned embedding models. My work combines deep learning, NLP, and time series forecasting, applying techniques like QLoRA fine-tuning, transformer internals analysis, and GloVe/TF-IDF embeddings to large-scale patent and publication data for innovation system research.

I'm also a PhD researcher in Machine Learning and the Physics of Complex Systems at the University of Brasília, where I apply maximum entropy network models and random walk dynamics to characterize the Brazilian innovation system. During my Master's, I built reservoir computing models from scratch — including the full mathematical derivation of Echo State Networks — to forecast COVID-19 time series. This mix of mathematical depth and applied engineering shapes how I work: I want to understand why a model behaves the way it does, not just get it into production.