Bridging the gap between complex data and actionable insights through Artificial Intelligence and Geoprocessing.

I am a Software Developer deeply committed to solving real-world problems using Machine Learning and Spatial Analytics. Recently, I've been focusing on Healthcare Efficiency, building end-to-end pipelines that integrate public health data with high-resolution demographic indicators.

I believe that programming is a form of art—a blend of logic, data, and social impact.

• Core: Python, R, Scikit-Learn, TensorFlow, PyTorch.
• MLOps & Lifecycle: MLflow (Experiment Tracking & Model Registry), Docker, Docker-compose.
• Data Processing: Pandas, Numpy, PyArrow, FastParquet.
• Feature Engineering: Advanced normalization, Log scaling, Census data integration.

• Python GIS: GeoPandas, geobr, PySAL.
• APIs: Sidrapy (IBGE Census 2022), PySUS (DATASUS FTP).
• ArcGIS Ecosystem: Pro, Online, Enterprise, Survey123, FieldMaps, ArcGIS API for Python.

• Automation: Airflow, GitHub Actions (CI/CD).
• Web/Mobile: TypeScript (React, Next.js), Dart (Flutter), PHP.
• Databases: PostgreSQL (PostGIS), MongoDB, MySQL, SQLite, Firebird.
• Cloud: AWS (Lambda, DynamoDB, API Gateway, S3).

Predicting Hospital Cost Gaps in Paraná (Brazil)

• The Problem: Identifying financial inefficiencies in chronic disease management (Diabetes/Hypertension).
• The Solution: An automated pipeline using PySUS and Sidrapy to fetch 2022 Census data, training a RandomForest model tracked by MLflow, and deploying a live inference dashboard via Streamlit.
• Key Achievement: Integrated diverse data sources to identify municipalities with >20% unexplained cost overruns through Residual Analysis.
• View Repository

• 🔭 Currently working on: Refining MLOps workflows and spatial anomaly detection.
• 🌱 Learning: Advanced CI/CD for ML (CML) and Deep Learning for satellite imagery.
• 💬 Ask me about: GIS integration, MLOps, or why Python is my Swiss Army knife.