Sen

About Sen

Highly driven computational scientist specializing in biomedical informatics, machine learning, advanced statistical modeling, and data visualization to solve complex problems in biomedical and diagnostic science Over 10 years of hands-on experience with C++, Unix/Linux, R, and Python, with deep expertise in building and optimizing production-grade NGS pipelines for whole-genome, exome, and RNA-Seq analysis Deep understanding of clinical laboratory environments, regulatory considerations, and diagnostic workflows, enabling accurate interpretation of complex biomedical data Hands-on experience designing and deploying cloud-native and AI solutions, leveraging serverless architectures (AWS Lambda) and developing LLM-enabled applications with Amazon Bedrock and LangChain Proven ability to collaborate across product, development, and testing teams to translate complex analytical findings into clear, actionable insights that drive technical decisions

Experience

1. Staff Bioinformatics Scientist

   Illumina

   Oct 2022 — Present · San Diego, CA, US

   Technical lead for the DRAGEN (C++ codebase) Germline SNV component, driving feature innovation and cross-functional collaboration across development and testing teams, resulting in a 23.4% improvement in germline calling accuracy over two software releases. Enabled clinical applications in population genomics, pharmacogenetics, and rare disease diagnostics Designed and implemented a structured error-classification framework to decompose accuracy gaps into high-impact categories, enabling prioritized root cause analysis and targeted solutions. This systematic approach led to a 16.8% overall reduction in germline SNV accuracy errors Led a hackathon initiative to build AI agents automating genomic workflows and failure diagnostics, enabling chatbot-guided assay selection, workflow execution (Nextflow), log-based error classification, and FP/FN root-cause analysis using Amazon Bedrock–powered LLM Mitigated systemic reference bias through the development of a haplotype-informed personalized reference pangenome, improving F1-score by 15.7% in hypervariable and clinically relevant genomic regions Key contributor to advancing somatic variant calling, enabling clinical applications in personalized treatment, minimal residual disease (MRD) monitoring, and early cancer detection

Education

• Arizona State University

  Doctor of Philosophy (Ph.D.), Biomedical Informatics

  2013 — 2015

• Arizona State University

  Master of Science (M.S.), Biomedical Informatics

  2011 — 2013

• Nanjing University

  BS, Biotechnology

  2005 — 2009