Our Team

Mahmoud is a first year PhD student focusing on different aspects of AI security including adversarial attacks and defenses for LLMs and ML systems.

Final year PhD student interested in sustainable AI systems. Her work focuses on linking inference characteristics with model's energy consumption.

Lewis is a senior PhD student and Founding Engineer at Mindgard, where he bridges academic research and industry practice in offensive AI security. His research explores how machine learning models and the systems built around them can be attacked and defended, with a particular interest in model extraction and circumventing safety guardrails. He is the author of Model Leeching and has contributed to several high-profile vulnerability disclosures affecting Nvidia, Microsoft, and Meta.

Zion is a first year PhD student in AI security, focusing on securing LLM agents and exploring how interpretability can inform defence mechanisms.

Maddie is a first-year PhD student working in AI security, focused on how we build systems that stay safe, reliable, and resilient when things get messy in the real world. Maddie is aiming for a career in security research, digging into the harder problems around robust and trustworthy AI. Outside the lab, you’ll usually find Maddie hiking, climbing, or out walking my giant dog—anything that gets her outdoors and off a screen for a bit.

William is a final-year PhD student and Founding ML Engineer at Mindgard. His work focuses on developing automated solutions for AI Red Teaming, targeting both individual models and complex AI systems. His research and industry contributions include the Automated Adversarial Assessment Framework (PINCH), methods for bypassing guardrail systems, and several high-profile security disclosures with Nvidia, Microsoft, and Meta.

Matthew is a final-year PhD student whose research focuses on intra-device scaling on networking hardware for national scale backbone infrastructure. His research is focused on control protocals for novel, scale-out topologies and their associated energy benifits against network reliability and quality of service.

Jie Liang is a first-year PhD student whose research interests focus on federated learning, edge caching, as well as the efficiency and interpretability of LLMs. He is dedicated to developing robust, secure intelligent distributed edge systems.

Hamid Nasiri is a Senior Research Associate in Sustainable Artificial Intelligence at Lancaster University, specialising in energy-efficient machine learning and parameter-efficient fine-tuning of large language models. He develops principled frameworks for scalable, resource-aware AI systems addressing real-world challenges across industrial, environmental, and healthcare domains. His work has been widely published in leading international journals and contributes to interdisciplinary collaborations on next-generation intelligent systems.

Daria is a PhD student at Lancaster University working on cost-aware tool orchestration for Agentic AI. While most frameworks prioritise performance, her research focuses on cost-aware decision making, aiming to make these systems efficient and sustainable.

PhD Student.

(2025) Postdoctoral researcher.

(2023) PhD graduate. Now Senior Research Engineering @ NPlan.

(2023) PhD graduate. Now Lecturer (Assistant Professor) @ University of Bristol.

(2023) Masters by Research graduate. Now PhD student @ City University of Hong Kong.

(2023) PhD graduate. Now Associate Senior Lecturer @ Uppsala University.

(2022) PhD graduate. Now Principal Software Engineering @ Huawei R&D UK.

(2020) PhD graduate. Now Associate Professor @ COMSATS.

(2019) Postdoctoral researcher. Now Assistant Professor @ Queen Mary University of London.