I'm a Computer Science student at PUCPR and MLops Engineer, focused on building end-to-end AI solutions — from research and modeling to production deployment.
Currently working on medical imaging, LLM agents, and predictive maintenance systems.
stack = {
"ml_frameworks": ["TensorFlow", "PyTorch", "Scikit-learn"],
"computer_vision": ["ViT", "Autoencoders", "CNN", "GNN"],
"llm_genai": ["RAG", "Fine-Tuning", "LangChain", "Agents", "LSTM"],
"time_series": ["Signal Processing", "Harmonic Analysis"],
"web": ["Node.js", "React", "Next.js", "TypeScript"],
"infra": ["PostgreSQL", "Docker", "GCP", "Git"],
"language": "Python (main)",
}Federated Learning for Kidney Stone Detection · IEEE SMC 2025 — Vienna Cross-domain medical imaging with privacy-preserving training (HIPAA). Published and presented internationally.
Unsupervised Autoencoders — Cross-Domain Generalization · PIBIC 2024–2025 STL model optimization for kidney stone detection. Accuracy improved from 40% → 73% on unseen datasets.
Vision Transformers in Digital Pathology · PIBIC 2025–Present Evaluating foundational ViT models for tumor classification and leprosy categorization in histopathology.
End-to-End Marketing Automation AI · Renault Transformation Day 2025 — 1st place Technical lead. Full automation pipeline: creative generation, publishing, and predictive KPI analysis.
- 1st place — Renault Transformation Day 2025
- Innovation Prize — Government of Paraná (OAB legal aid automation project)
- International publication — IEEE SMC 2025, Vienna
B.Sc. Computer Science · PUCPR · Expected Dec 2027