A collection of multi-part blog series exploring technical topics in depth.
LLM Comprehension
Measuring what a language model actually does with the text you give it, one method per post.
Programming with Invariants
Moving input rules into the values themselves, checked once at the boundary, so the rest of the code can rely on them.
Research Passes
Surveys of a research and engineering field before deciding to build in it.
Optimizing SAMURAI
Taking a video segmentation model from 1.3 FPS to real-time.
Inside Gradient Boosting
How modern gradient boosting libraries like XGBoost and LightGBM work, from first principles to implementation details.