I Wrote a GPU Matmul Kernel From Scratch in Triton. Here's Everything I LearnedI recently started learning Triton, OpenAI's Python-based language for writing GPU kernels. My project: build a matrix multiplication kernel from scratch, step by step, until it's competitive with PyT15 min read·galacodes.hashnode.devJun 23
GPUs: The Hardware That Power AIGPUs revolutionized AI with parallel computing, transforming video game technology for models like ChatGPT and Claude18 min read·galacodes.hashnode.devNov 30, 2025
Speculative Decoding: From Theory to ImplementationThe complete guide to understanding the concept of speculative decoding in LLM inference and implementing it from scratch15 min read·galacodes.hashnode.dev1DNov 5, 2025
Bits Don't Lie: Datatypes in Modern LLMsLet’s talk about the different datatypes that are being used in modern LLMs like GPT, LLaMa and the like. The most common ones that you might have heard: FP32, BFloat16, Float16, INT8, etc. These are all standard data types and available in PyTorch a...6 min read·galacodes.hashnode.devMay 30, 2025
Introduction to LLM inferencingA complete introduction to LLM Inference with KV Cache and sample code.7 min read·galacodes.hashnode.devOct 26, 2024