Shuji Ohira

Introduction

I am a Security Consultant at Amazon Web Services Japan, where I help customers design and implement secure cloud architectures with a focus on identity and access management. Before joining AWS, I earned my Ph.D. in Computer Science from Nara Institute of Science and Technology (NAIST), where I was a member of the Internet Architecture and Systems Laboratory (inet-lab). During my doctoral studies, I was also a Research Fellow of the Japan Society for the Promotion of Science (DC2).

My research focused on cybersecurity for cyber-physical systems, particularly in-vehicle networks such as Controller Area Network. I explored both offensive and defensive approaches, examining how physical-layer characteristics (e.g., voltage and timing anomalies), entropy-based indicators, and kernel-level signals can be leveraged to design—or evade—intrusion detection systems.

After completing my Ph.D., I joined AWS as a Support Engineer, where I worked closely with customers to troubleshoot complex issues and developed deep technical expertise in AWS security services. This experience gave me a practical perspective on how large-scale distributed systems behave in real-world environments.

Across both research and industry, I’ve been particularly interested in the boundaries of trust in complex systems—how assumptions about identity, communication, and protocol behavior can break down in unexpected ways. Today, I continue to explore these themes in cloud and identity systems, especially protocols such as SAML and OpenID Connect, where cross-domain trust relationships can create subtle security challenges.