PhD Researcher in Computer Science @ University of Manchester
LLM Safety · Formal Verification · Multimodal Learning · Anomaly Detection
My PhD research focuses on LLM safety, with a current emphasis on developing formal verification methods and guarantees for the performance and reliability of large language models. Previously, during my MSc I worked on contrastive vision-language models, with a particular focus on data- and parameter-efficient fine-tuning of vision-language foundation models for zero-shot image anomaly detection. Some of this work resulted in Crane and TIPS Over Tricks.
Embedded Systems & Low-Level Development
Before moving into deep learning research, I spent years during my BSc passionately working with embedded systems, low-level software, and modern C++. An outcome of this work is WAVRecorder, a user-friendly open-source library for audio recording.
