London

Research Scientist.Next: MPhil, Cambridge.

I’m especially interested in AI safety: mechanistic interpretability, alignment, emergent misalignment, and model persona selection.

Selected work

[1]

2026 / joint first author

We wrote a public, editable constitution for AI systems. Read the paper.

[2]

2026 / arXiv

I worked on an LLM! Read the paper.

2026–2027 Cambridge: MPhil, Advanced Computer Science
University of Cambridge
Incoming student at St John’s College.
2021–2025 Manchester: BSc, Computer Science
The University of Manchester
First class / 4.0 GPA.