I am a Artificial Intelligence Researcher, Backend Developer, and PhD Candidate in Data Science at the Aeronautics Institute of Technology (ITA). My work sits at the intersection of Data-Centric AI, Generative Modeling, and Healthcare, investigating how Large Language Models and Diffusion Models can solve missing data challenges across tabular datasets and medical imaging.
- PhD in Data Science — Aeronautics Institute of Technology (ITA / PPG-PO)
[2025 – Present]- Visiting PhD Researcher (PDSE/CAPES Scholarship) — Portugal (Generative AI & Medical Image Reconstruction)
- MSc in Operations Research & Data Science — ITA & UNIFESP
- 🏆 1st Place at the CTD / CSBC (Brazilian Computer Society) Thesis & Dissertation Contest
- 🥉 3rd Place at the Brazilian Symposium on Operations Research (SBPO)
- BSc in Medical Physics — UFCSPA
- Missing Data Imputation
- Data-Centric AI
- Generative AI
- Medical Image Reconstruction
- Responsible AI
- Languages: Python, R, .NET
- ML/DL: PyTorch, TensorFlow, scikit-learn, XGBoost, TabPFN
- Data Science: pandas, NumPy, matplotlib, seaborn
- Tools: Git, GitHub, Jupyter, VS Code, LaTeX, Overleaf
TRAXI: An Explainable Retrieval-Augmented Framework for LLM-Based Tabular Missing Data Imputation.
- Leverages RAG and Large Language Models (LLMs) to impute missing tabular attributes conditioned on relevant context retrieved from observed data.
- Built on top of the scikit-learn API architecture (
fit/transform) with native support for explainability metrics.
🔹 mdatagen
A Python library for synthetic missing-data generation under realistic mechanisms.
- Generates controlled missingness benchmarks under MCAR, MAR, and MNAR mechanisms for robust tabular evaluation.
- 🌐 Collaborative research, PhD exchange inquiries, or preprint discussions: LinkedIn or Email