Comprehensive Guide
AI InfrastructureFrom Silicon to Intelligence
A systematic, in-depth reference covering the entire AI infrastructure stack — from GPU kernels and CUDA to distributed training, inference optimization, and production platforms.
219 chapters published across 8 modules
The Guide
8 modules,219 chapters.
Each module builds on the layers below — from hardware foundations up to production applications.
Blog
Latest posts & deep dives.
Quantization in LLM Inference: Techniques, Tradeoffs, and What's Next
A 70B model in FP16 takes 140 GB. Your GPU has 80 GB. Your laptop has 16. Quantization is the most important practical technique for closing that gap — not as a compromise, but as an engineered tradeoff. This post walks through how it actually works, from block quantization math to dp4a kernels to edge deployment.
The Omni Model: How a Tiny LLM Learns to Listen, See, Think, and Speak
A visual deep-dive into MiniMind-O — a 0.1B-parameter Omni model that processes text, audio, and images as input, and generates text and streaming speech simultaneously. Covering the Thinker-Talker architecture, Multi-Token Prediction for audio codebooks, training data anatomy, and the complete evolution from text LLM to multimodal Omni.
From Text to Vision: What Changes When an LLM Learns to See
A visual walkthrough of how a text-only LLM becomes a vision-language model — the vision encoder, projector, token replacement trick, attention patterns on image tokens, and the two-stage training strategy. Using real data from a 160M-parameter VLM.
Start from the foundations.
Begin with the transformer architecture and hardware fundamentals, then work your way through CUDA, distributed systems, training, inference, and production platforms.