My research focuses on the design of system software, spanning OS, hypervisor, and trusted execution environment (TEEs). I aim to build observable and trustworthy foundations for cloud infrastructure, such as SaaS, FaaS, and agentic AI platforms.

Currently, I work on enhancing confidential computing architectures, particularly confidential virtual machines (CVMs) and also user-land library operating systems (LibOS). My goal is to extend CVM isolation for sandboxes and confidential containers, improving the security, reliability, and efficiency of cloud infrastructures and AI code agents.