vLLM: LLM Inference as a memory and scheduling problem

Part 1 of a practical deep dive into vLLM: prefill vs decode, the KV-cache bottleneck, PagedAttention, and how modern attention backends consume paged KV.

A Brief History of Computer Vision: Before Pixels Had Brains

Part 1 of N tracing the journey of computer vision.

Why the Hell Am I Writing a Blog?

A quick intro to why I started this blog

Hello world!

Notion Test Blog