Senior Machine Learning Engineer building production ML and GenAI systems in healthcare - Bayesian modeling, causal inference, and LLM applications (RAG, agentic workflows, fine-tuning) that clinicians actually use.
- 🩺 Senior ML Engineer @ TapestryHealth - I lead ML/AI initiatives end-to-end: Bayesian causal time-series models for patient-vitals monitoring, a RAG-based clinical question-answering system serving ~6,000 requests/month on Ray, real-time PHI de-identification with distilled transformers, and agentic document-processing workflows.
- 🔬 Core developer of PyMC - the open-source probabilistic programming library for Bayesian modeling in Python.
- 📄 Published researcher - peer-reviewed work in ML and statistics
I write deep-dive posts implementing ML foundations from scratch at my blog, Advanced Analytics & Engineering:
- Transformer architectures, attention, and tokenization - implemented in JAX
- Word embeddings (Word2Vec, GloVe) and contextual representations (ELMo, BERT)
- LLM pretraining and instruction tuning
- Production ML engineering with Ray, MLflow, and Kedro
- Bayesian spatial-temporal modeling
Python · JAX · PyTorch · PyMC · Hugging Face · LangChain/LangGraph · Ray · Spark · scikit-learn · XGBoost · MLflow · Airflow · Docker