Software engineer with a decade in software engineering and HealthTech, now specializing in Generative AI and production-ready MLOps. With deep healthcare domain experience and two peer-reviewed publications, my focus is building scalable, reliable AI systems that solve real-world problems. I thrive on moving complex models from prototype to deployment, with an emphasis on mitigating hallucination, model interpretability, and test-driven development.
- Leveraging Community Health Workers for Predicting Emergency Department Readmissions — International Journal of Semantic Computing, Vol. 18, No. 01, April 2024
- Harnessing Machine Learning to Optimize Community Health Worker Interventions for High-Risk Patients — Computers and Their Applications, Springer Nature Switzerland, 2026
- Cross-Site Generalization in Otoscopy Deep Learning — Manuscript in preparation (co-first author)
| Project | Description | Key Technologies |
|---|---|---|
| vertex-care | Operationalized my peer-reviewed research on patient readmission by building an agentic MLOps platform that generates actionable intervention plans for high-risk patients. | Google Gemini, ReAct Agent, LLM Feature Extraction, FastAPI, GCP Cloud Run, CI/CD |
| scholar-agent | An advanced multi-agent research assistant using LangGraph, a Knowledge Graph (Neo4j), and a re-ranking RAG pipeline to reason over scientific papers. | LangGraph, Google Gemini, Neo4j, ChromaDB, sentence-transformers |
| claim-triage-ai | An intelligent healthcare claim triage pipeline that predicts denials, clusters root causes via NLP, and routes high-priority claims for automated processing. | XGBoost, Sentence-BERT, Clustering, FastAPI, Streamlit, Docker, CI/CD |
| Cross-Site Otoscopy Domain Adaptation | A deep-learning study on unsupervised domain adaptation for cross-site medical imaging — adapting a ResNet50 classifier across institutional datasets (DANN/MCC/CORAL) with occlusion-based interpretability analysis. (co-first author; manuscript in preparation) | PyTorch, ResNet50, Domain Adaptation, Interpretability |
