At Illumina, I built an LLM-powered research assistant using RAG and Python to analyze genomic literature and scientific reports, with 6,000 research hours and $450K in projected annual labor savings. I also developed NLP workflows to extract genetic variants, disease associations, and clinical evidence from scientific literature.
I’ve developed predictive models and RNA-seq workflows for genomic research, and built XGBoost ranking models to prioritize disease-related variants, reducing manual review time by 35%. Earlier at Illumina, I built NGS pipelines for quality control, read alignment, and variant calling, and automated genomic data cleaning and validation using Python and R.

