I am a Computer Science student specializing in Artificial Intelligence and Machine Learning, with experience developing and deploying ML models using frameworks such as PyTorch and TensorFlow. I have expertise in Machine Learning system design and model optimization techniques, including weight pruning and structural pruning.
I have implemented neural network architectures including Convolutional Neural Networks (CNNs), Multilayer Perceptrons (MLPs), 3D Convolutional Networks (Conv3D), Transformers, and Vision Transformers (ViTs), delivering solutions for image, text, and multimodal data.
I also have experience in deploying and integrating AI models through cloud platforms (Microsoft Azure), as well as via API requests, leveraging tools such as Cognitive Services Pipelines, Form Recognizer, Azure Automated Machine Learning, Azure Designer, RAG workflows, and AI-powered chatbots to deliver scalable, production-ready solutions.
|
Object Detection
Wheat Head Detection using Knowledge Distillation
YOLOv5, Knowledge Distillation, MLflow, Databricks, Pytorch, Albumentations. |
2D Image Regression/Object Detection
VisionGauge: A Computer Vision Model to Detect and Read U-Tube Manometers
Transfer Learning, Albumentations, PyTorch, CNN, Residual Blocks, Finetunning, ETL, OpenCV |
|
2D Image Classification
Sugarcane Leaf Disease Predictor
FCNN, CBAM, Xception, Residual Blocks, TensorFlow, OpenCV |
3D Image Classification
RSNA Intracranial Aneurysm Detection
Transfer Learning, 3D CNN, Multilabel Classification, K-Fold, Keras, OpenCV |
|
MLP Binary Classification
Modelo de Classificação Binária para Previsão de Retorno ao Presencial
Fully Connected Neural Network, Dropout, ReLU, TensorFlow, Keras |
|
Santos, C. S. dos, & Arima, M. N. (2026). VisionGauge: a computer vision model to detect and read U-tube manometers. Artefactum - Revista De Estudos Interdisciplinares, v. 25 (n. 5), E3462, Qualis A3. DOI: 10.23900 |
Santos, C. S. dos, Arima, M. N., & Almeida, M. A. D. (2025). Development of a pipeline network simulator and experimental validation on a test bench. Caderno Pedagógico, v. 22 (n. 12), E21100, Qualis A2. DOI: 10.54033 |
|
CONTROLE DE ESTUDOS
Software para Controle de Estudos
Python, Kivy, KivyMD, HTML, TeX, kvlang |
CONTROLE DE ESTOQUE
Software para Controle de Estoque
Python, Kivy, KivyMD, HTML, TeX, kvlang |
- Banco de Dados E-commerce