Hi, I'm Kingsley Kim. I'm a current student at UVA, mostly interested in kernel optimization and LLM research. I've worked under Chen-Yu Wei on RL applied to control problems and LLMs, as well as training more efficient Process Reward Models. I also worked on a more applied paper on Transformers in Time Series.

Outside of school, in Summer 2025 I worked at Palantir as a Forward Deployed Engineer in Manufacturing.

I've also written a few posts on CUDA and LLM inference, and some notes on textbooks I've read: PRML, Sutton's RL.

Contact: