I'm an undergraduate student pursuing a Bachelor of Science and Technology at Ilum โ School of Science, with interests in Data Science, Machine Learning, and Artificial Intelligence. I work as an undergraduate researcher in the Data Science Group at the Brazilian Synchrotron Light Laboratory (LNLS/CNPEM), where I develop and evaluate self-supervised learning methods and foundation models for 3D representation learning in synchrotron X-ray tomography.
My broader interests include Computer Vision, Physics-Informed Machine Learning (PINNs), and Large Language Models (LLMs), particularly in interdisciplinary scientific and bioinformatics applications.
| ๐๏ธ |
Data Science Intern ยท LNLS / CNPEM Evaluating Neural Network architectures (U-Net, Swin Transformer) for semantic segmentation of synchrotron microtomography volumes. Main stack: PyTorch, Python, NumPy. |
| ๐ |
B.Sc. in Science and Technology ยท Ilum โ School of Science Interdisciplinary program with emphasis on scientific programming, mathematical modeling, and data analysis. |
| Project | Description | Status |
|---|---|---|
| Ilum | Unified academic repository โ coursework and projects throughout my B.Sc. | |
| srcPINN | Source term identifiability in parabolic PDEs via Physics-Informed Neural Networks (PyTorch) | |
| SwinCheX | Chest X-ray pathology classification using Swin Vision Transformer | |
| ฮ-XTB | Delta-ML correction over GFN2-xTB energies to approximate DFT results โ ML final project (2025.2) | |
| Langevin | Numerical simulation of stochastic dynamics via Langevin equations โ ODE final project (2025.2) | |
| SIR Model | Epidemiological modeling fitted to real COVID-19 data โ Data Science final project (2025.1) | |
| SVD | Singular Value Decomposition applied to image compression and noise reduction โ Linear Algebra final project (2025.1) |
๐ก Scientific imaging & computer vision ๐งฎ Differential equations & numerical simulation
๐ค Scientific machine learning (SciML) ๐ฌ Synchrotron experimental data analysis
๐ Mathematical modeling & symbolic computation ๐ง Neural networks for physical problems (PINNs)