Shunfan (Shelven) Zhou

Lead Researcher, Phala Cloud
Ph.D. in Software Engineering

I am the lead researcher of Phala Network, an open-source and trustless computing cloud powered by Trusted Execution Environments, co-founded in 2019. We are pioneers in TEE GPU technology and provide TEE GPU support for secure AI workloads. We offer the open-source SDK dstack that enables applications to run in TEE with no modifications and easy verifiability.

I received my B.A. in Computer Science from Fudan University. I received my Ph.D. in Software Engineering from Fudan University in 2022, jointly advised by Prof. Min Yang and Prof. Peng Liu (Penn State University).

My research interests include TEE, program analysis, and distributed systems architecture. I am currently exploring confidential AI via TEE.

Open to collaborations: I welcome partnerships with researchers working on confidential computing, secure AI, and privacy-preserving machine learning.

Selected Publications

Confidential Computing on NVIDIA Hopper GPUs: A Performance Benchmark Study
arXiv preprint, 2024
Performance analysis showing minimal TEE overhead for most AI workloads
State-Aware Symbolic Execution for Exploring State-Dependent Program Paths
USENIX Security 2021
Novel program analysis technique achieving higher code coverage
An Ever-evolving Game: Evaluation of Real-world Attacks and Defenses in Ethereum Ecosystem
USENIX Security 2020
Comprehensive analysis of blockchain security landscape

Recent Blog Posts

Dstack: A Zero Trust Framework for Confidential Containers
2025-05-23
Verifying TEE Applications: A Practical Walkthrough with Dstack
2025-06-19
Securing Domain Certificates: Ensuring Exclusive Control by TEE
2024-11
Key Management Protocol For Decentralized Root-of-Trust
2024-10