Return to Phuong M. Cao's home page
Phuong Cao is a scientist at the National Center for Supercomputing Applications (NCSA) at the University of Illinois at Urbana-Champaign (Illinois). He is a U.S. National Science Foundation (NSF) Cybersecurity Center of Excellence, Trusted Cyberinfrastructure (Trusted CI) Fellow, an affiliate of the Center for Artificial Intelligence Innovation at Illinois, and a member of the Joint Laboratory for Extreme Scale Computing (JLESC). Before joining Illinois, he gained a broad security perspective at Microsoft Research, IBM Research, LinkedIn, and Akamai. He was a visiting researcher at the Pohang University of Science and Technology (POSTECH), a member of FPT Young Talents Technology Centre (FYT), VNG, and Bach Khoa Internet Security (BKIS). He received his bachelor's degree from Hanoi University of Science and Technology.
His research focuses on securing AI for science: building supercomputers that are resilient against novel adversaries (AI-, quantum-driven cyberattacks) and system-wide outages induced by hardware uncertainty and software incorrectness. He is leading a cross-stack research portfolio addressing all aspects of the dependability of systems and high-speed interconnects. Examples of awarded grants include: post-quantum cryptography migration, security data lake, GPU error analysis on sustained petascale supercomputers/hybrid-cloud/DeltaAI, verification of federated authentication, and research security policies. His research has been published in top-tier conferences, including USENIX Security, USENIX Networked Systems Design and Implementation (NSDI), ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis (Supercomputing), and IEEE Quantum Computing and Engineering (QCE). His probabilistic graphical models for failure prediction or attack preempting are grounded in field data from real-world workloads and security incidents. He maintains close collaboration with research computing centers (San Diego Supercomputer Center, Pittsburgh Supercomputing Center, Rice University Center for Research Computing, and the National Energy Research Scientific Computing Center) and national labs (Oak Ridge National Laboratory and the Stanford Linear Accelerator Center).
The impact of his research are being realized in Department of Energy's leadership computing facilities, such as Argonne National Lab, and adopted in FABRIC, an NSF-funded, international, programmable, and mid-scale distributed research infrastructure. He has given tutorials on security and forensics at major universities, facilities, and labs, including Carnegie Mellon University, the National Center for Atmospheric Research, Lawrence Berkeley National Laboratory, and the National Institute of Standards and Technology (NIST). His research has received a Best Paper Award at the 44th IEEE/IFIP International Conference on Dependable Systems and Networks (DSN), been showcased at the Colorado Convention Center in Denver, and been featured in HPCwire. His Art of HPC on network attacks has been exhibited at the Georgia World Congress Center. He is working closely with Illinois School of Information Sciences, ECE, Siebel School of Computing and Data Science to bring real-world examples into classes such as Probability with Engineering Applications, Dependable AI systems, and Computer Security.
Successful students mentored by and collaborated with him have taken critical positions at NVIDIA, Apple, xAI, Google, NASA Jet Propulsion Laboratory, Sandia National Labs, and have been awarded the Jerry Fiddler Innovation Fellowship. He has guided students to win several hackathons, including Box.com (first prize), Salesforce.com, and the Steven Ashby Prize in Computational Science (honorable mention). For his service, he has been recognized as an Outstanding Mentor for CyberCorps®: Scholarship for Service students.