About
ML research engineer at Renaissance Philanthropy. I build AI products and do mechanistic safety research.
Interests
- AI safety. Mechanistic interpretability; a model's internal thought process, what it says versus what it thinks; probing and steering the residual stream.
- Intuition. Making beautiful things, discovering my unknown unknowns, energetic connection with the universe.
- Art. Music channeling, jams, production.
Recent updates
- Knowing Before Saying accepted at MATH-AI, NeurIPS 2026.
- GeoGenBench talk at AIME-Con; poster at MATH-AI.
- On Metaphysical Connection with the AI, talk at The Science of Consciousness. Slides
- Applying to PhD programs.
Education
- MASc, Systems Design Engineering, University of Waterloo.
- BHSc, Health Sciences, McMaster University.