Biography
My lifestyle is guided by analytical thinking, curiosity, and the development of consistent definitions. My professional interests also include computational sciences, mathematics, and the implementation of holistic learning systems. If any of the topics I’ve mentioned or my research piques your interest, please feel free to reach out!
Research interests
AI Alignment aims to steer AI systems toward a person's or group's intended goals, preferences, or ethical principles. Mechanistic Interpretability is about reverse engineering neural networks to decompose them into (hopefully) human-understandable components like features and circuits in order to increase our understanding and steer the models towards desirable outcomes during inference.
My current research focuses on understanding mechanistic interpretability as a field, how Sparse Auto Encoders (SAEs) can isolate features from a neural networks and how mechanistic interpretability can be combined with AI Alignment in order to create more safe and robust AI models.