Healthcare technology and medical-imaging professional building reliable software, data, and AI systems for real clinical workflows.
I bring more than 16 years of PET/CT, Nuclear Medicine, and Computed Tomography experience to software engineering and data/AI engineering. My work starts with the operational or clinical problem, then applies reproducible engineering, careful validation, and responsible-AI boundaries.
- PETQuant Reliability — safety-oriented PET reliability logic, data-quality contracts, and false-reassurance analysis.
- Chest CT Classification — medical-imaging transfer-learning audit with deterministic evaluation and explicit clinical limits.
- Pima Diabetes ML Evaluation — leakage-safe healthcare ML with missing-measurement handling and clinically relevant metrics.
- Responsible AI — SHAP + LIME explainability, fairness evaluation, and adversarial robustness.
- AWS Glue + EMR PySpark Ingestion Validation — catalog resolution, schema validation, and fail-closed ingestion-quality checks.
Explore the ten-project curated portfolio →
I build healthcare technology through OraDigit and RadVita, with current work spanning medical-imaging analytics, clinical workflow tools, responsible AI, mobile/software engineering, and secure Firebase-backed systems. Product source and operational infrastructure remain private; the public portfolio above is the evidence-focused view of selected engineering methods.
Python · Dart/Flutter · JavaScript/Node.js · SQL · Firebase · AWS/PySpark · scikit-learn · TensorFlow/PyTorch · testing and CI
- 16+ years in PET/CT, Nuclear Medicine, and CT
- Bachelor of Science in Data Science
- Graduate studies in Software Engineering
- Founder, OraDigit
- Start with the clinical or operational problem.
- Make provenance, assumptions, and limitations visible.
- Test data contracts and failure paths—not only happy paths.
- Treat AI outputs as evidence to evaluate, not authority to trust blindly.
- Build technology that supports healthcare professionals rather than replacing their judgment.