Guanqin ZHANG

I am a Postdoctoral Research Associate in the School of Computer Science and Engineering at the University of New South Wales, working with Professor Yulei Sui. My current research centers on Responsible AI and agentic workflows for building trustworthy, verifiable AI systems. I am the main designer and developer of SVF-tools/ACT, an end-to-end neural network verification and optimization toolkit built on the SVF framework.

Previously, I earned my PhD at UNSW (2022–2025) within CSIRO’s Data61 Responsible AI program, with my thesis Order-Leading Branch-and-Bound for Neural Network Verification, focused on verifying deep neural networks and improving their robustness. During my PhD I was also guided by Dr. Dilum Bandara and Dr. Shiping Chen at Data61.

My overarching objective is to close the gap between legacy, conventional workflows and an automated, dynamic, and trustworthy cycle.

Beyond my formal research, I am a self-confessed enthusiast of large language models (LLMs) — I love building agentic tools and pushing LLM-driven coding and research workflows to their limits in my day-to-day work, as the token counter below attests.