MS Analytics (Computational Data Track) — Georgia Tech · BS Cognitive Science (ML & Neurocomputational Focus) — UCSD

Data engineering + ML for longitudinal/biomedical data. I build reproducible pipelines, research-grade analyses, and clean model evaluation.

View repos · LinkedIn · Google Scholar

• Wearable data pipeline — Postgres ingestion → dbt transforms/tests → CI via GitHub Actions
  → https://github.com/mrkchoe/wearable-data-pipeline
• Model comparison / evaluation — reproducible baselines + metrics + reporting
  → https://github.com/mrkchoe/ad-mri-pet-model-comparison
• Sanctuary operations data platform — operational data modeling and SQL-based reporting for intake, care, adoption, and cost analytics in a sanctuary environment
  → https://github.com/mrkchoe/sanctuary-operations-data-platform
• Airflow API → PostgreSQL ELT pipeline — Dockerized Apache Airflow DAG orchestrating API ingestion, structured transformation, PostgreSQL loading, and basic data validation in a reproducible local environment
  → https://github.com/mrkchoe/airflow-api-to-postgres-demo
• Neuroimaging publications — curated bibliographic list of peer-reviewed articles with structured authorship context supporting longitudinal MRI/biomarker research
  → https://github.com/mrkchoe/publications

Languages
Python · SQL

Data Engineering & Infrastructure
PostgreSQL · Apache Airflow · dbt (models, tests, documentation)
Docker · GitHub Actions
Relational schema design · Data validation · Reproducible batch pipelines

Machine Learning & Evaluation
scikit-learn · Feature preprocessing
Cross-validation frameworks · Model evaluation diagnostics (ROC, PR, confusion matrices)
Structured model comparison & research-grade reporting

Visualization
D3.js (interactive statistical visualization)
Seaborn (statistical plots & model diagnostics)

Java / R / Bash
AWS (EC2, S3, SageMaker)
TensorFlow / Pytorch / Neural network modeling
Longitudinal biomedical dataset management