Jeremias Lino Ferrao

Reinforcement Learning and Mechanistic Interpretability Researcher

My research tackles the challenge of open-endedness by understanding the mechanisms of learning and exploration. I believe that building more capable agents and making them interpretable are mutually reinforcing goals; progress in one accelerates the other. This dual pursuit is motivated by a scientific curiosity in the principles of discovery and the practical need for safe, aligned AI.

• Email
• Google Scholar
• GitHub
• LinkedIn
• Twitter
• Hugging Face