Hi! I'm Andrei Alexandru. I'm passionate about making the future go well for humans, especially in light of increasing AI capabilities.

I work on AI safety research, focusing on evaluating large language models' capabilities and understanding the alignment problem.