---

I am an Assistant Professor in the Computer Science Department at UCLA. I am a systems builder interested broadly in system security and networked systems. My work has been recognized with several awards, including a Jay Lepreau Best Paper Award, an IRTF Applied Networking Research Prize, and an ACM SIGSAC Doctoral Dissertation Award Runner-Up. My PhD research on TCPlp significantly influenced the Thread network standard, developed by a consortium of companies in the IoT space (including Google/Nest, Apple, Qualcomm, etc.), and has been adopted in OpenThread, an open-source netowrk stack used in some of these companies' products. My CV is available here (last updated in August 2026).

Prior to joining UCLA, I spent a year as a postdoctoral scholar in the FOCI Center at the University of Washington, supervised by Arvind Krishnamurthy and Ratul Mahajan. Before that, I received my Ph.D. in Computer Science in the EECS Department at UC Berkeley. I was fortunate to be co-advised by David E. Culler and Raluca Ada Popa. During my Ph.D., I worked in the Sky Computing Lab, the RISE Lab (the predecessor to the Sky Computing Lab), and the Buildings, Energy, and Transportation Systems (BETS) research group. Prior to that, I received my B.S. in Electrical Engineering and Computer Sciences at UC Berkeley. As an undergraduate student, I worked in the Software-Defined Buildings (SDB) research group, which was the predecessor to BETS.

In Fall 2026, I will be teaching CS 239 (course website coming soon).

Research Interests

I am interested broadly in system security and networked systems. I am especially interested in building systems that achieve security or efficiency properties, or that navigate the tension between security and efficiency. Lately, I have been investigating the following research directions:

• Privacy-preserving data analytics and AI. How can users benefit from data analytics and AI while keeping their data private? How can parties (e.g., organizations, like hospitals or banks) combine their datasets for data analytics or AI, while keeping their datasets private from each other?

  Recent work: Osprey (OSDI 2026). See also: MAGE (OSDI 2021, Best Paper Award).

• Efficient datacenter systems. How can we make cloud and datacenter workloads (e.g., distributed applications organized as microservices, or GPU-intensive AI applications) more efficient, while preserving isolation among different tenants' or users' applications?

  Recent work: GVM (conditionally accepted to ASPLOS 2027). See also: Skyplane (NSDI 2023).

• Trustworthy and dependable agentic AI. How can we enable users to benefit from agentic AI while minimizing the risks associated with its unreliability or untrustworthiness (e.g., susceptibility to prompt injections)?

  Recent work: NaSh (technical report).

These descriptions are high-level—if you are at UCLA or interested in collaborating, then feel free to ask me about the details.

Students

PhD students:

• Sai Achalla (co-advised with Remy Wang)
• Yicheng Liu (co-advised with Harry Xu)
• Sebastian Jiang (co-advised with Konstantinos Kallas)
• Robert Stanley

Masters students:

• Yutong Zhou (M.S. 2026)

In my research group, I am working with several undergraduate students: Avi Verma, David Sun, and Woyu Wang. I also work with Nakul Khambhati (B.S. 2025, now a PhD student at UCLA advised by Rafail Ostrovsky), Joonwon Lee (B.S. 2025), and Gary Song (B.S. 2025, now a PhD student at Georgia Tech).