Sharan Maiya

I’m a first year PhD candidate based in the Language Technology Lab at the University of Cambridge, where my work mostly focuses on interpretability methods and evals with language models.

I’m also interested in climate change mitigation, and I'm affiliated with CamCID and climaTRACES. I’m supervised by Anna Korhonen and Ramit Debnath and funded through the AI4ER CDT.

My current broad research-goals are to build AI-based tools to accelerate science, specially in the domains of the physical sciences and environmental risk. I'm interested in representation learning, interpretability and language models. In a previous life I worked on causal discovery algorithms and targeted causal learning, looking at the health effects of exposure to air pollution.

I also spend a lot of time thinking about meditation, AI safety, whatever movie I watched yesterday, and how much I miss Scottish tap water, so reach out if you want to talk about any of those (or perhaps about research, I guess). You can do that at firstname98lastname [at] gmail.com or through the links below.