Nitin Madnani is a distinguished research engineer in the Research & Development division at ETS. He received his Ph.D. in computer science in 2010 and his M.S. in computer engineering in 2004—both from the University of Maryland, College Park. He received his B.E. (honors) in electrical engineering from Punjab Engineering College, Punjab University (India).
He is a senior member of the NLP Architecture team in the Natural Language Processing (NLP) AI lab. He is one of the original architects of Rater Feature Services (RFS), a robust, low-latency, and fault-tolerant service-based framework used as the basis of almost all of the scoring and feedback engines developed by ETS. In addition, he is the lead developer and maintainer for SKLL, ETS’s machine learning library, and RSMTool, ETS’s automated scoring evaluation library, both available as open-source Python packages. He is also the current technical lead for Relay Reader™, ETS’s collaborative reading app designed to help developing readers improve their oral reading fluency. In the past, Madnani has worked on automated essay scoring, content scoring, plagiarism and canned-response detection, and building automated feedback systems for writing.
He is also an active member of the broader NLP community. He currently serves as chief information officer and member-at-large on the Association for Computational Linguistics (ACL) executive committee and as an executive member of the ACL Special Interest Group for Building Educational Applications (SIGEDU). Over the years, he has produced more than 60 publications including journal articles, book chapters, and conference & workshop papers. He has also served as program committee member, area chair, and senior area chair for several international NLP conferences. He is currently an action editor for Transactions of the Association for Computational Linguistics. Most recently, he was invited to co-author a book on automated essay scoring by Morgan Claypool as part of their Synthesis Lectures on Human Language Technologies series.