PhD Researcher in Healthcare AI · Building fair, calibrated ML for Type-2 Diabetes risk prediction
I'm a PhD researcher at the University of Portsmouth developing machine learning models for the early detection of Type-2 Diabetes risk in South Asian populations. My work focuses on fairness, calibration, and explainability in clinical risk prediction — the technical conditions under which a model is safe to deploy in diverse populations.
The research is directly relevant to populations with very high T2D prevalence, particularly the GCC region and South Asia, where adult diabetes prevalence is among the highest globally and existing risk scores often perform worst.
Alongside the doctoral research, I build production ML systems in healthcare AI. I'm currently looking for Machine Learning Engineer, AI Engineer, and Applied Scientist roles — UAE-based or remote-friendly.
A public, productionised ML pipeline for fair, calibrated, explainable Type-2 Diabetes risk prediction. Subgroup audits with Fairlearn, SHAP-based explainability, FastAPI deployment, MLflow tracking.
A production-grade retrieval-augmented Q&A system over NICE NG28 and ADA Standards of Care for diabetes, with hybrid retrieval, hallucination evaluation, and Langfuse tracing.
The fastest way to reach me is email. I'm particularly interested in conversations about fairness in clinical ML, Type-2 Diabetes risk prediction, or healthcare AI in high-prevalence populations.