A technical newsletter on machine learning systems, covering GPU performance, PyTorch internals and kernel-level behavior, written by a PyTorch engineer. It also includes personal career reflections and posts on new APIs such as FlexAttention.
Website: https://www.thonking.ai
A Korean-language engineering blog by a search engine developer covering GPU architecture, vector compression research and personal updates. Readers get technical explainers written from a practitioner's perspective, plus occasional reflections on work and study.
A personal blog series covering CUDA GPU programming, from optimizing single-precision matrix multiplication to low-precision arithmetic in ML and profiling kernels. Readers get tutorial-style explanations of GPU performance concepts, with code mostly pointed to external GitHub and Medium links.
One short email a week with hand-picked newsletters from the index. Free, no spam, unsubscribe anytime.
