FDA Fellowship - Language Models; LLM in Drug Safety Reviews

Description

FDA‑CDER offers a research fellow position within the Office of Pharmacovigilance and Epidemiology (OPE) and Office of Surveillance and Epidemiology (OSE). The fellow will focus on evaluating large‑language‑model (LLM) performance for literature review tasks such as abstract and title screening and information extraction. This research is part of the Center for Drug Evaluation and Research (CDER) mission to ensure safe and effective drugs.

• Expand LLM benchmark datasets using FDA reviewers’ literature reviews.
• Develop prompts and design benchmark and validation studies to assess additional literature review tasks.
• Research multi‑step processes and agentic models, comparing them with single‑step chatbot models and evaluating their methodologies.

• Master’s or doctoral degree in a relevant field, earned within the last five years or to be earned by 31 August 2026.
• Profound knowledge of epidemiological principles and experience with medical literature review.
• General knowledge of LLMs and prompt design; experience with LLM architecture or local API usage is a plus.

• U.S. citizenship, lawful permanent residency, or other eligible immigration status as described in the program guidelines.
• Proof of health insurance and a successful background investigation by the Office of Personnel Management.
• Compliance with FDA ethics requirements for non‑employee scientists.

For questions, contact Ashley .CDER. Reference code: FDA‑CDER‑2026‑0100.

8/14/2026 3:00 PM Eastern Time Zone.