Cong Ye

Cong Ye‘s primary responsibilities include study design, sampling design, data analytics, and psychometric analysis. He has extensive experience in experimental design, sampling design, weighting for complex designs, sampling error estimation, imputation, small area estimation, and analysis of complex survey data. He has also led the development of open-source fine-tuned large language models, AI agents, data analytics tools, complex databases, and web applications at AIR. He is certified Project Management Professional (PMP) and Google Cloud Certified Professional Cloud Architect.

Dr. Ye has provided guidance on experimental design, imputation, sampling, weighting, and statistical modeling for the U.S. Department of Education's large-scale complex surveys and various health surveys. He developed an open-source sampling package, RSZ, in Stata; led psychometric and scoring development for the ED School Climate Surveys (EDSCLS) and created algorithms to calculate Rasch scale scores on the fly using R in the EDSCLS survey platform; guided small area estimation methodologies for the National Household Education Survey and the California Market Rate Survey; led the development of data analytics tools including topic modeling, sentiment analysis, web scraping, video processing, and text labeling using Python; built a data warehouse and dashboard that integrates dozens of data sources across five geographic levels to support researchers and practitioners in generating insights from linked data; developed web applications to streamline literature review and manage item banks; fine-tuned a BERT model that has been successfully applied to text classification projects on related topics; and led the development of an AI agent for general data analysis.