A researcher with opinions about models.

I'm working on the practical side of language models — the part where they have to actually do something useful for someone.

I'm doing a PhD at MBZUAI in Abu Dhabi under the gentle supervision of Tim Baldwin. My research is about quantifying and measuring THE AUDACITY of LLMs — how confidently they fabricate, overclaim, and otherwise behave like they own the place.

Outside research I read too much, take too many photos of plants, and try to make websites that feel like rooms rather than dashboards.