Portfolio
Projects
Selected case studies and projects across applied AI, healthcare, MLOps, and data products.
Featured case studies
Cases with explicit context, decision, and outcome for a more editorial reading path.
- Case study
Featured
Fator Noronha — what does it cost to live on an archipelago?
How a partnership between UPE and the Fernando de Noronha Territorial Authority mobilized applied econometrics and computational modeling to decompose price differentials between the archipelago and mainland Brazil — and what it taught about applying data science where the problem is public policy, not product.
Context
Fernando de Noronha has a chronically high cost of living compared to the mainland. The Territorial Authority and UPE needed a rigorous diagnosis: how much of the differential is unavoidable logistical cost, and how much is avoidable inefficiency — and where to intervene first.Decision
Structure the research as an econometric decomposition problem: build a representative price basket, model the determinants of the differential by component (transport, scale, taxation, market structure) and simulate logistical intervention scenarios with computational modeling.Outcome
Ongoing research funded by the Territorial Authority via PROPEGI/UPE, producing structured diagnostics that feed public policy decisions on island logistics and supply. - Case study
Featured
Clinical Decision Support for Heart Failure: from guidelines to code
How a medication triage system for hypertension and heart failure was built by translating AHA clinical guidelines into structured knowledge bases, predictive models, and a production API — and what that process taught about the gap between medical knowledge and working software.
Context
Vitally Health builds clinical decision support tools for cardiologists and care managers. The challenge: medication titration for HFrEF/HFpEF patients is complex, guideline-driven, and highly individualized — but the guidelines are written for clinicians, not for code.Decision
Design a hybrid rule-and-model system that encodes AHA guidelines as structured business rules, uses ML to rank medication titration options, and serves decisions via a FastAPI + AWS Lambda endpoint.Outcome
A production clinical decision support API integrating predictive analytics into the care platform, with medication ranking that balances personalization, safety, and guideline compliance.
Projects
Additional work that rounds out the portfolio.
Metametrics — Computer Vision for Quality Control
Computer Vision / IndustryTech lead for a startup building real-time process monitoring systems using computer vision, developed as part of CESAR School's startup acceleration program.
PythonOpenCVPyTorchUltralytics (YOLO)SimpleCPT — Automated Clinical Coding
Health AI / NLPAutomated ICD-10 coding suggestion system for Brazilian electronic health records, built with pre-trained Transformer networks and a rank-based learning framework.
PythonPyTorchHuggingFace TransformersScikit-learn