
I am an AI security researcher at RAND Center on AI, Security, and Technology, where I serve as a principal investigator on ASI robustness and AI control research, focusing on the application of information security principles, cryptography, complexity theory, and formal methods to questions of high-assurance control and enforcement in advanced AI systems.
I received my Ph.D. in Computer Science from Harvard University, advised by Prof. Stephen Chong and Prof. Srinivas Devadas at MIT CSAIL. During my PhD, I spent time at Microsoft Research and Cloudflare Research, where I conducted research in systems security and confidential computing.
Email | Curriculum Vitae | Github | Harvard | Office Hours
Rethinking Runtime Integrity Guarantees in Distributed AI Frameworks. (Under Submission).
Varun Gandhi, Simon Langowski, Stephen Chong, and Srinivas Devadas
Rethinking Verifiable Execution for Confidential AI Compute
Varun Gandhi, Stephen Chong, and Srinivas Devadas
Rethinking System Audit Architectures for High Event Coverage and Synchronous Log Availability
V. Gandhi, S. Banerjee, A. Agarwal, A. Ahmad, S. Lee, and M. Peinado
32nd USENIX Security Symposium, 2023
[Paper]
[Slides]
[Code]
[Talk]
Automated recovery of far edge computing infrastructure in a 5g network
S. Sariou, V. Gandhi, A. Wolman, and L.P. Cox
U.S. Patent 11,900,127
[Paper]
Liveness guarantees in secure enclaves using health tickets
S. Sariou, V. Gandhi, A. Wolman, and L.P. Cox
U.S. Patent 12,067,111
[Paper]
Rethinking Isolation Mechanisms for Datacenter Multitenancy
V. Gandhi, and J. Mickens
12th USENIX Workshop on Hot Topics in Cloud Computing, 2020
[Paper]
[Slides]
[Talk]
Shadow PC, IEEE Security and Privacy, 2021
Teaching Fellow, CS 153 Compilers, Fall 2023
Teaching Fellow, CS 262 Introduction to Distributed Computing, Spring 2023
Teaching Fellow, APCOMP 221, Critical Thinking in Data Science, Spring 2020