Hi, I’m Amaan.
I am a data scientist at a leading health provider, building machine learning that supports clinical decisions.
Curiosity
Asking how a system works, then building the tools to find out. It is how I started programming, and how I still begin every problem.
Expression
Making a result clear enough for someone else to act on. A model is only as useful as the explanation that comes with it.
Mastery
Preparing thoroughly, being tested, and improving with each round. Rigour is a habit before it is a method.
Foresight
Thinking several steps ahead: how a system will behave on data it has not seen, and what happens when it is wrong.
Meaning 2021 – 22
Finding structure in language. At Siemens, multilingual customer comments clustered by intent. At SRM, 250,000 research abstracts organised into topics, recognised with a Best Paper award.
Proof 2023 – 25
Evidence before conclusions. Graduate study in statistics at Rutgers, causal inference in faculty research, and validated graph neural networks for drug discovery at Rutgers New Jersey Medical School.
Purpose 2026 – now
Applying both to patient care. At Virtua Health, I build clinical risk models and language pipelines that help care teams act earlier.
How I work
- Measure honestly. Held-out data, temporal validation, and the metric that matches the cost of being wrong.
- Explain the result. If a clinician cannot see why a patient was flagged, the model is not finished.
- Build for use. Models are deployed, monitored and maintained, not left in a notebook.
- Keep it efficient. Lean systems are cheaper to run and reach more people.