Sierra — AI Researcher · Sep 2024 – Present · San Francisco, CA. Lead the τ-Bench family of agent benchmarks (research direction, codebase, public leaderboard at taubench.com, and a sequence of extensions: τ²-Bench, τ-Knowledge, τ-Voice, τ³-Bench — three papers accepted to ICML 2026). Kickstarted Sierra’s internal agent evaluation system and the user-simulation-driven stress testing used to harden agents pre-release.
Elemental Cognition — Senior Researcher (NLP) · Jun 2023 – Jul 2024 · Remote. Core algorithms for EC’s hybrid AI architecture integrating LLMs with formal reasoning. Led the multi-round-RAG + causal-modeling work published as From Generating Answers to Building Explanations at NAACL 2025 (Industry).
Uniphore.ai — Principal NLP Scientist & NLP Research Lead (previously Senior NLP Scientist) · Dec 2019 – Jun 2023 · Palo Alto, CA. Built and led Uniphore’s NLP Lab (7+ scientists); set short-, mid-, and long-term research strategy for understanding human–human spoken conversation in the contact-center domain.
Astound.ai — AI Scientist (NLP) · Sep 2017 – Dec 2019 · Menlo Park, CA. Core member of the team that built Astound’s domain-expert conversational agent for IT / HR / Finance enterprise self-service from the ground up.

University of Southern California — PhD, Computational Cognitive Neuroscience · 2010 – 2017. Dissertation: Schema Architecture for Language–Vision Interactions. Advisor: Michael A. Arbib.
École Normale Supérieure, Paris — MS, Cognitive Science · 2008 – 2010.
École Polytechnique, Palaiseau — MS, Physics · 2003 – 2006.
Lycée Louis-le-Grand, Paris — Classes Préparatoires, Mathematics & Physics · 2001 – 2003.

English (fluent) · French (native) · Spanish (conversational).

The PDF below is the canonical, formatted version — see it for full publication and selected-works lists.