Location: New York, NY
Role Summary
Develop and evaluate machine learning and GenAI models to drive clinical, operational, or financial insights from healthcare data.
Responsibilities
- Train supervised and unsupervised models using Python (XGBoost, LightGBM, sklearn, PyTorch)
- Conduct data profiling, feature engineering, model evaluation using stratified validation
- Implement model explainability (SHAP, LIME) and bias audits
- Monitor model drift, fairness, and clinical relevance over time
- Fine-tune or prompt LLMs for use cases such as summarization or information extraction
- Build and evaluate multi-agent workflows using GenAI and frameworks like LangChain
- Design prompt libraries and benchmark hallucination rates, sensitivity to input phrasing
- Track experiments using MLflow, Weights & Biases
Required Qualifications
- 3 5 years in applied data science or machine learning
- Experience with Snowflake SQL, Python ML ecosystem, healthcare data
Preferred Qualifications
- Exposure to Epic Clarity data or unstructured clinical notes
- Hands-on with transformer models, LLM APIs, and embedding-based retrieval
