Hands-on Build Guide: Tiled Self-Attention Kernel with Online Softmax in Pure NumPy
A step‑by‑step NumPy implementation of tiled soft‑max attention, mirroring FlashAttention’s efficiency, perfect for a CV‑boosting side project.
Performance
Hands‑On Build Guide: Speculative Decoding Engine in Pure Python
A step‑by‑step guide to creating a speculative decoding prototype in pure Python, showcasing algorithmic design, performance tuning, and production‑ready patterns.
Optimizing NVIDIA CUDA Graphs for Sparse Matrix Multiplication on GPU Clusters
This post details the end-to-end optimization of CUDA Graphs for sparse matrix multiplication across GPU clusters, covering architecture, implementation patterns, and performance benchmarks.
Pure-Python INT8 Weight Quantizer for Transformer Inference: A Hands-On Build Guide
Learn to implement a production‑ready INT8 quantizer with per‑channel scaling, test it on transformer weights, and extend it for real‑world deployment.
Mastering eBPF for Kernel-Level Tracing: Techniques and Real-World Patterns
Learn how to harness eBPF programs to safely instrument the kernel, collect metrics, and build low‑overhead tracing pipelines.