Noam Benkler

Mr. Benkler is a SIFT research scientist specializing in Artificial Intelligence. Mr. Benkler's work spans multi-agent LLM systems, physiological signal processing, and NLP. At SIFT, Mr. Benkler is presently developing physiology-driven crew-monitoring capabilities for the DARPA VANESSA program, and a multi-agent software-porting system that autonomously adapts C/C++ software to CHERI-enabled architectures under the DARPA SWITCH program. At SIFT, Mr. Benkler has also spearheaded machine-learning research in computational social science, leveraging small and large language models to construct population-level attitude networks and assess trends in complex cultural beliefs under DARPA's HABITUS program. He developed methods for measuring the moral and cultural values expressed in text; work that produced his Recognizing Value Resonance (RVR) models and first-author publications at the NeurIPS Moral Value Pluralism workshop (2023), LREC-COLING (2024), and SBP-BRiMS (2022). His research has also modeled team dynamics through multilevel physiological profiling and the prediction of team strategic dissent from wearable data.

Mr. Benkler holds an M.Sc. in Data Science from Tufts University, which he completed while working full-time at SIFT, and a B.A. in Economics from Carleton College.