~/archive

Everything, newest first — 119 posts since 2024.

2026

03-30 What Happens to FlashAttention When Vera Rubin Ships? #mlsys/hardware
03-30 FlashAttention-4: When Tensor Cores Got Too Fast for Everything Else #mlsys/hardware
02-23 Spark, Cerebras, and the Future of Low-Latency AI Inference #mlsys/hardware
02-15 MLIR Is Not Just Another IR #mlsys/compiler
01-19 vLLM and PagedAttention: Why KV Cache Management Matters #mlsys/inference
01-16 LLM Serving 101: Prefill, Decode, Batching, and the Systems Behind Large Language Models #mlsys/inference

2025

09-20 MXFP4 in GPT-OSS : Why Everyone Talks About It #mlsys/quantization
06-06 Context-Free Grammar (CFG) #mlsys/compiler
06-04 Backus Naur Form (BNF) #mlsys/compiler
06-02 Compiler Basic Structure #mlsys/compiler
05-31 Postfix Notation #mlsys/compiler
05-07 LLVM-Flow (OSS) #mlsys/llvm
05-05 LLVM-Block (OSS) #mlsys/llvm
04-29 History of LLVM #mlsys/llvm
04-27 Debug & Metadata #mlsys/llvm
04-26 Optimization Pass #mlsys/llvm
04-25 Static Single Assignment (SSA) #mlsys/llvm
04-24 Basic Block & CFG #mlsys/llvm
04-22 LLVM IR Syntax #mlsys/llvm
04-21 LLVM IR #mlsys/llvm
04-20 LLVM Basic Structure #mlsys/llvm
02-13 Host-Device Synchronization #mlsys/cuda
02-09 Kernel Configuration #mlsys/cuda
02-05 Advanced Atomic Operations #mlsys/cuda
01-30 Basic Atomic Operations #mlsys/cuda
01-25 Bank Conflict #mlsys/cuda
01-20 Memory Alignment and Coalescing #mlsys/cuda
01-17 Thread Hierarchy #mlsys/cuda
01-15 CPU vs. GPU Architecture #mlsys/cuda
01-12 Intro to CUDA #mlsys/cuda

2024

10-10 GPT-1 #ml/papers
10-03 Transformer #ml/papers
09-23 Evaluation in NLP #ml/transformer
09-20 Transformer #ml/transformer
09-12 Tokenizing #ml/transformer
09-10 Self Attention #ml/transformer
09-06 Seq2Seq #ml/transformer
09-04 RNN and Attention #ml/transformer
09-01 RNN and its cons #ml/transformer
09-01 RNN and its pros #ml/transformer
08-15 EfficientNet #ml/papers
08-11 MobileNet V3 #ml/papers
08-07 MobileNet V2 #ml/papers
08-03 SENet #ml/papers
07-30 MobileNet V1 #ml/papers
07-26 ResNeXt #ml/papers
07-24 DenseNet #ml/papers
07-20 WideResNet #ml/papers
07-18 Pre-Act. ResNet #ml/papers
07-15 Inception V2/V3 #ml/papers
07-12 ResNet #ml/papers
07-10 Inception V1 #ml/papers
07-07 VGGNet #ml/papers
07-03 R-CNN #ml/papers
05-27 mAP #ml/object_detection
05-26 Soft-NMS #ml/object_detection
05-25 IoU & NMS #ml/object_detection
05-24 Selective Search #ml/object_detection
05-22 Sliding Window #ml/object_detection
05-21 Concepts of Object Detection #ml/object_detection
05-16 Attention #ml/dl_theory
05-15 Seq2Seq #ml/dl_theory
05-14 RNN Back Propagation #ml/dl_theory
05-14 RNN #ml/dl_theory
05-13 CNN Summary #ml/dl_theory
05-12 VGGNet #ml/dl_theory
05-11 CNN Feature Map #ml/dl_theory
05-10 Padding, Pooling, Striding #ml/dl_theory
05-09 CNN with 3D input #ml/dl_theory
05-09 CNN Feature Extraction #ml/dl_theory
05-08 CNN #ml/dl_theory
05-07 Regularization #ml/dl_theory
05-06 Dropout #ml/dl_theory
05-06 Overfitting & Data Augmentation #ml/dl_theory
05-05 Loss Landscape / Skip-Connection #ml/dl_theory
05-03 Batch Normalization #ml/dl_theory
05-02 Vanishing Gradient #ml/dl_theory
05-02 ReLU #ml/dl_theory
05-01 Universal Approximation Theorem #ml/dl_theory
04-30 Multi-class Classification #ml/dl_theory
04-29 Neural Network is MLE machine #ml/dl_theory
04-28 MSE vs. Likelihood #ml/dl_theory
04-27 Perceptron #ml/dl_theory
04-27 Binary Classification (Sigmoid) #ml/dl_theory
04-26 Back Propagation with Matrix Derivatives #ml/dl_theory
04-25 Back Propagation #ml/dl_theory
04-24 MLP & Non-Linearity #ml/dl_theory
04-23 K-fold Cross Validation #ml/dl_theory
04-23 Train, Valid, Test #ml/dl_theory
04-22 Adam #ml/dl_theory
04-22 Momentum vs. RMSProp #ml/dl_theory
04-21 Mini-Batch SGD #ml/dl_theory
04-21 Stochastic Gradient Descent #ml/dl_theory
04-20 Weight Initialization #ml/dl_theory
04-20 Gradient Descent #ml/dl_theory
04-19 Linear Regression #ml/dl_theory
04-19 Neural Network #ml/dl_theory
04-13 Information Theory #ml/math
04-12 Maximum A Posteriori #ml/math
04-12 Maximum Likelihood Estimation #ml/math
04-11 Uniform and Normal Distribution #ml/math
04-11 Mean and Variance #ml/math
04-11 Random Variable and Probability Distribution #ml/math
04-10 Vector derivative respect to Matrix #ml/math
04-10 Matrix derivative respect to Matrix #ml/math
04-09 Scalar derivative respect to Matrix #ml/math
04-08 Vector derivative respect to vector (Chain Rule) #ml/math
04-08 Vector derivative respect to vector #ml/math
04-07 Gradients #ml/math
04-06 Scalar derivative respect to vector #ml/math
04-05 MacClaurin & Taylor Expansion #ml/math
04-04 Partial Derivative #ml/math
04-04 Chain Rule #ml/math
04-04 Derivative #ml/math
04-03 Limits #ml/math
04-02 Transpose & Dot Product #ml/math
04-02 Vector & Matrix #ml/math
04-01 Logarithmic Function #ml/math
04-01 Function #ml/math