Noah D. Goodman

Also known as Noah Goodman

Noah D. Goodman is a professor of psychology and computer science at Stanford whose work on human cognition and AI includes the Church and Pyro probabilistic programming languages and the STaR reasoning method. After mathematics degrees at Arizona and Texas, he joined MIT in 2005 and was first author of the 2008 paper introducing Church. At Stanford since 2010, he runs the Computation and Cognition Lab and was named a Sloan Research Fellow in Neuroscience in 2016; he worked on Pyro at Uber in 2018, coauthored STaR in 2022, and became a research scientist at Google DeepMind in 2024. In January 2026, humans&, an AI lab he co-founded, raised a $480 million seed round at a $4.48 billion valuation.

Now at: Stanford University
Badges: Founder
Bridge

Timeline

1997–2003

Mathematics at Arizona and Texas

He earns a BA in mathematics and a BS in physics at the University of Arizona in 1997. He completes a PhD in mathematics at the University of Texas at Austin in 2003.
2005–2010

Church at MIT

He joins MIT as a postdoctoral associate in 2005 and becomes a research scientist there in 2008. In 2008, he is first author of "Church: a language for generative models" with Vikash Mansinghka, Daniel Roy, Keith Bonawitz and Joshua Tenenbaum. Church is a family of Lisp-like probabilistic programming languages whose rules of inference are probabilistic.
2010–2016

Stanford and a Sloan Fellowship

He moves to Stanford in 2010 as an assistant professor of psychology, with courtesy appointments in computer science and linguistics, and runs the Computation and Cognition Lab. In 2016, he is named an Alfred P. Sloan Research Fellow in Neuroscience.
2018

Pyro at Uber

In March 2018, he talks at Berkeley's RISE Lab about Pyro, a universal probabilistic programming language written in Python on a PyTorch backend, which he has been building at Uber. In October 2018, he is among the authors of "Pyro: Deep Universal Probabilistic Programming", which describes stochastic variational inference built on PyTorch.
2022

STaR bootstraps reasoning

He coauthors "STaR: Bootstrapping Reasoning With Reasoning" in 2022 with Eric Zelikman, Yuhuai Wu and Jesse Mu. The Self-Taught Reasoner has a language model generate rationales, fine-tune on those that yield correct answers, and repeat, and it performs comparably to fine-tuning a 30 times larger model on CommonsenseQA.
2024–2025

DeepMind and the Singapore Consensus

From 2024, he is also a research scientist at Google DeepMind, focusing on AI safety and language model post-training. In April 2025, he is a contributor at the Singapore Conference on AI, and in June 2025 he is among the authors of the Singapore Consensus on Global AI Safety Research Priorities, the conference's final report.
January 2026

Co-founding humans&

In January 2026, the AI lab humans&, whose founders include him, raises a $480 million seed round at a $4.48 billion valuation. He describes its aim as a humanist AI built around collaboration, trust and shared outcomes.
As of October 2026

Professor at Stanford

He is a Professor of Psychology and of Computer Science at Stanford, a faculty affiliate of the Institute for Human-Centered Artificial Intelligence, and a member of the Wu Tsai Neurosciences Institute. He teaches Minds and Machines in autumn 2026, AI Agents and Simulations in winter, and a spring course on whether language models have mental states.

Career

Since 2010Stanford UniversityAssistant Professor of Psychology
From 2026humans&Co-founder
From 2024Google DeepMindResearch Scientist
2005–2010Massachusetts Institute of TechnologyPostdoctoral Associate, Research Scientist