My postdoc work focused on vulnerability data quality, automated repair, and deception detection. My work focused on how robust label error detection is to near-duplicate and tampered samples. My work focused on automated repair methods for software vulnerabilities. My work focused on deception detection that accounts for figures of speech. My Ph.D. applied deep reinforcement learning to combinatorial optimization in transportation.

Amutheezan Sivagnanam is a Data Engineer at Shell. Previously, he was a Postdoctoral Fellow at the Reasoning and Data Analytics for Security (ReDAS) Lab at the University of Houston, where his work focused on software vulnerability dataset quality, label error detection robustness, automated software vulnerability repair, and figure-of-speech-aware deception detection.

He earned his Ph.D. in Informatics from The Pennsylvania State University in Summer 2025, where he was a member of the Applied Artificial Intelligence Lab advised by Dr. Aron Laszka. His doctoral research focused on applying artificial intelligence to solve combinatorial optimization problems in transportation domains. He also earned a Master’s degree in Computer Science from the University of Houston in Summer 2022, and completed his Bachelor’s degree in Computer Science and Engineering at the University of Moratuwa, Sri Lanka, in 2017.

Before his doctoral studies, he gained industry experience as a Software Engineering Intern at WSO2 Lanka (Pvt) Ltd, working with the Data Analytics Team from Jul to Dec 2016. He later worked as a Software Engineer at LSEG Technology (formerly MillenniumIT Software (Pvt) Ltd), contributing to the Post-Trade Team from Jan 2018 to Jul 2019.