Blog
My thoughts, experiences, and insights on development, technology, and more.
Stages of a program in Cuda
A walkthrough of how a CUDA kernel launch progresses through grid dispatch, warp scheduling, execution, and retirement on the GPU's streaming multiprocessors.
Paper Review: ActiveLLM - Large Language Model-Based Active Learning for Textual Few-Shot Scenarios
A comprehensive review of the ActiveLLM paper exploring how instruction-tuned LLMs can revolutionize active learning by selecting high-impact training samples without cold-start problems or iterative retraining delays.