Sanjeev Arora
Sanjeev Arora is a theoretical computer scientist whose research connects computational complexity with machine learning and AI reasoning. He earned degrees at MIT and Berkeley, joined Princeton in 1994, and led a center for computational intractability. His honors include the Packard Fellowship, the 2011 ACM-Infosys Foundation Award, and ACM SIGACT’s 2024 Distinguished Service Award. At the Institute for Advanced Study, he led a theoretical machine-learning program; Princeton launched Princeton Language and Intelligence under his direction in 2023, and he discussed how language models combine skills in May 2025. As of October 2026, he is on leave from his Princeton professorship and is founding director of Princeton Language and Intelligence.
- Now at
- Princeton University
- Works
- 349
- Citations
- 22,918
- h-index
- 68
Publication numbers from OpenAlex, as of 15 October 2026.
Timeline
-
1990–1994
MIT and Berkeley
He earns an SB in Mathematics with Computer Science from MIT in 1990 and a PhD in Computer Science from Berkeley in 1994. In 1994, he joins Princeton as a faculty member.
-
1997
A Packard Fellowship
The Packard Foundation names him a fellow in computer and information sciences in 1997. His work focuses on theoretical computer science, machine learning, and algorithms with provable guarantees.
-
2008–2013
A center and an ACM award
He becomes founding director and lead PI of Princeton’s NSF-funded Center for Computational Intractability in 2008 and serves through 2013. The ACM and Infosys Foundation give him the 2011 ACM-Infosys Foundation Award in the Computing Sciences for innovative approaches to problem solving.
-
2017–2020
Theoretical machine learning at IAS
From 2017 through 2020, he divides his working time between Princeton’s computer science department and the Institute for Advanced Study, leading a new program in theoretical machine learning.
-
2023
Princeton Language and Intelligence
Princeton launches Princeton Language and Intelligence in September 2023 under his direction. He leads efforts to explain the mathematical underpinnings of large language models.
-
2024
A SIGACT service award
ACM SIGACT presents him with its Distinguished Service Award in 2024 for advocacy that increases funding and visibility for theoretical computer science, and for leading STOC’s transformation into TheoryFest.
-
May 2025
How language models learn
In May 2025, he discusses how language models combine abstract skills and show signs of metacognition.
-
As of October 2026
Professor and PLI director
In March 2026, he presents Gödel Prover V2 for Lean-based theorem proving and a new inference pipeline that reaches state-of-the-art performance on IMO-ProofBench (Advanced) at the time of evaluation. As of October 2026, he is on leave from his Charles C. Fitzmorris professorship in Computer Science and is the founding director of Princeton Language and Intelligence.
Sources and credits
Photo: We have no published photo of Sanjeev Arora yet. Send us a free-licensed photo.
- NextG 2026 Symposium Speakers | Princeton Engineering
- Sanjeev Arora
- Arora, Sanjeev • The David and Lucile Packard Foundation
- Arora honored for insights in working around the impossible - Princeton Engineering
- Diving into the Mysteries of Deep Learning | CS
- AI at Princeton: Pushing limits, accelerating discovery and serving humanity | Princeton Alumni
- 2024 ACM-SIGACT Distinguished Service Award
- Sanjeev Arora on Parrots No More: How AI Models Learn, Reason, and Self-Improve - Bendheim Center for Finance
- How could a Superhuman AI mathematician come about? | Math