I am deeply passionate about bridging the gap between theoretical interpretability and applied safety in clinical AI. Operating largely as an independent researcher outside of traditional lab structures, I have driven my own projects from investigating causal auditing via activation steering to publishing sole-authored empirical work. My long-term goal is to contribute to the AI safety ecosystem by building tools that guarantee foundation models are aligned, auditable, and genuinely reasoning rather than just pattern-matching. Beyond research, I am an intermediate competitive chess player, which fuels my appreciation for rigorous, forward-looking strategic evaluation.