Post-Doctoral Fellow in Statistical Analysis and Tool Development for GenAI Evaluation & Watermarking Metrology National Institute of Standards and Technology in United States

Post-Doctoral Fellow opportunity in Statistical Analysis and Tool Development for GenAI Evaluation & Watermarking Metrology at the National Institute of Standards and Technology (NIST), supported through the George Washington University and the NIST Professional Research Experience Program (PREP).

This position focuses on the statistical backbone of NIST’s GenAI evaluation work, including the development of NIST-CARAT (Calibrated Risk Assessment Tool for authentication technologies). The role is centered on statistics, Bayesian modeling, uncertainty quantification, ROC/EER/Brier score analysis, and Bayes-risk characterization for evaluating watermark robustness and content quality trade-offs across text, image, voice, video, and code.

The post emphasizes building reproducible, containerized data pipelines and analytical tools that can support policy and standards work related to AI provenance, deepfake detection, and watermark compliance thresholds. The work also includes translating statistical findings into white papers and technical submissions for international standards bodies such as ISO/IEC.

Eligibility highlights: a Ph.D. in Statistics, Biostatistics, or a closely related highly quantitative field; strong experience with experimental design and classification metrics; proficiency in R or Python; shell scripting; Docker; workflow tools such as Argo Workflows; and dashboard/interactive visualization tools such as R Shiny.

Funding: the post is a postdoctoral fellowship/research appointment through NIST PREP, but no stipend or salary amount is stated in the post.

Location: NIST Campus, Gaithersburg, Maryland, United States.

Application: use the provided application link to apply. No deadline is stated in the post.