M.S. in Computer Science (Machine Learning) from Georgia Tech.
I build research software and AI-enabled systems, with a focus on backend infrastructure, evaluation, healthcare and behavioral data, and reliable software for research workflows.
My work spans FastAPI/React applications, data and ML pipelines, containerized services, observability, and reproducible research systems.
Full-stack RAG system with FastAPI, Weaviate, Docker Compose, Prometheus/Grafana observability, health checks, CI/CD, and user feedback.
Synthetic-data healthcare-AI research prototype (FastAPI, Next.js, SQLite). Implemented hybrid retrieval, then measured it on a frozen 20-query / 49-chunk fixture: query expansion hurt, MiniLM/hybrid underperformed BM25, and the demo default followed that evidence. Corpus provenance documents the 15/14 source split. Not for clinical use; does not claim HIPAA compliance.
Modular behavioral-data research pipeline for Reddit collection, LLM-assisted classification, sentiment analysis, topic modeling, statistical analysis, and reproducible evaluation.
Clinical decision-support prototype integrating FastAPI, SMART on FHIR, structured EHR data, and LLM-assisted triage.
Backend & Infrastructure
Python · FastAPI · SQL · PostgreSQL · Docker · GitHub Actions ·
Prometheus · Grafana · Redis · REST APIs
Research & Data
pandas · NumPy · PyTorch · scikit-learn · statistical evaluation ·
experiment tracking · behavioral and healthcare data
AI / ML
LLMs · RAG · model evaluation · Hugging Face · PyTorch ·
retrieval systems · NLP
Frontend
TypeScript · React · Next.js
- Scientific and research software
- Healthcare and behavioral AI
- Reliable AI evaluation
- Human-centered data systems
- Research infrastructure and tooling