data scientist · ML researcher · MSc candidate

I turn messy data into
decisions people trust.

I'm Sam — a data science grad student working on interpretable machine learning for healthcare. I like models that explain themselves and plots that need no caption.

Explore my work Get in touch
0research projects
0publications
0%best model AUC
0TBdata wrangled

Projects that shipped

model AUC vs baseline

Sepsis early-warning model

Gradient-boosted survival model predicting sepsis onset 6h earlier than the clinical baseline, deployed in a hospital pilot.

XGBoostSHAPsurvival analysis
weekly active users after launch

Campus energy dashboard

Live dashboard forecasting energy use across 40 university buildings; saved an estimated 8% on heating costs.

ProphetAirflowGrafana
0% accuracy gain

Fine-tuned compact language models for medical triage notes — 85% smaller, same accuracy.

TriageLM

Distilled clinical language model for triage classification, running on-prem for privacy.

PyTorchdistillationNLP
bike demand vs temperature

City bike-share rebalancing

Demand forecasting + optimization that cut "no bikes available" complaints by a third in simulation.

scikit-learnOR-ToolsGeoPandas

Research journey

MSc Data Science · thesis on interpretable ML

Building faithfulness metrics for post-hoc explanations in clinical models.

Research intern · National Health Data Lab

Federated learning pipelines across six hospitals without moving a single patient record.

BSc Statistics · first-class honours

Thesis: Bayesian hierarchical models for small-area disease mapping.

Publications

2026

Do explanations help clinicians? A randomized study of SHAP in triage

Okafor S., et al. — under review

2025

FedTriage: privacy-preserving triage models across hospital silos

Okafor S., Lindqvist A. — ML4H Workshop, NeurIPS

2025

Calibration drift in deployed clinical models: a 12-month audit

Okafor S., et al. — JAMIA Open

2024

Bayesian small-area estimation of respiratory disease burden

Okafor S. — undergraduate thesis, awarded department prize

Toolbox

PythonRSQL PyTorchscikit-learnXGBoost pandasSparkAirflow dbtDockerTableau

Let's collaborate

Open to research collaborations, internships and interesting datasets. The weirder the data, the better.

sam@example.com
⚡ Built with FolioForge — get yours from $5/mo