Yiming Jia

About Yiming Jia

Hello! I’m a data science enthusiast currently pursuing my M.S. in Data Science at New York University, building on a strong foundation in Applied Mathematics and Data Science from UC San Diego. I am currently seeking 2026 new grad full time opportunities in data science and AI, with an expected graduation in May 2026.My expertise lies in predictive modeling, data analysis, and machine learning, where I thrive on transforming complex data into actionable insights that drive impactful business outcomes. I’m passionate about using data to optimize strategies and enhance efficiency, always aiming to bridge the gap between technical solutions and real world challenges. In my current role as an Applied AI Data Scientist Intern, I design LLM powered reasoning modules and structured prompts that turn unstructured application data and documents into reliable, policy aligned eligibility decisions.Beyond the numbers, I love connecting with people, sharing insights, and learning from their stories and experiences.Outside of work, I’m an explorer. Whether it’s traveling to new places, capturing moments through photography, playing the guitar, experimenting in the kitchen, hiking, or exploring coffee shops, my interests keep me grounded, curious, and open to new experiences. These passions fuel my creativity and approach to problem solving.I’m driven, curious, and always eager to learn and take on new challenges. Let’s connect!

Experience

1. Data Scientist

   Genmab

   Sep 2025 — Present · NJ, US

   Built a scalable, parameterized pharmacovigilance analytics pipeline integrating FAERS, EudraVigilance, and JADER to standardize CRS safety analysis across oncology drugs, applying the workflow to Epcoritamab CRS monitoring and supporting cohort level, repeatable risk reporting with 344 CRS cases identified from - Developed explainable severity prediction models on highly imbalanced FAERS oncology data, using PR-AUC as a primary evaluation signal and achieving ROC AUC 0.66 and PR AUC 0.44, then generating SHAP and LIME explanations to support risk stratification of rare fatal outcomes beyond frequency based monitoring- Designed an automated signal detection framework combining Isolation Forest anomaly detection, FDA label filtering, and disproportionality testing, analyzing 58,296 FAERS reports and flagging 1,386 rare and unexpected drug adverse event signals validated by PRR, IC025, and chi-square thresholds to enable focused, earlier safety signal review.

Education

• UC San Diego

  Bachelor of Science - BS, Applied Mathematics

• Grace Christian School

  High School Diploma

• New York University

  Master of Science - MS, Data Science