Skip to the document
Madhuopen lab
Outcome School · Inference and serving6 min read

by Amit Shekhar · 14 June 2026

LLM Inference Optimization

Techniques like KV Cache, Paged Attention, Flash Attention, Speculative Decoding, Continuous Batching, and Prompt Caching are what make LLMs fast and scalable in production.

1,167 words#llm#ai#system-design0 recall cards

LLM Inference Optimization
Read on Outcome School ↗then come back to lock it in

What this article covers

  1. KV Cache in LLMs
  2. Paged Attention in LLMs
  3. Decoding Flash Attention in LLMs
  4. Grouped Query Attention
  5. Speculative Decoding
  6. Continuous Batching in LLMs
  7. Prompt Caching
  8. Prefill vs Decode: LLM Inference Optimization
  9. How does vLLM work?
  10. How does SGLang work?

The article lives on outcomeschool.com. Read it there, then come back: the tutor in the margin has read it and will answer questions, and the questions below check what stayed.

Before you go

In one sentence, what was this chapter about?

From memory, without scrolling up. Writing it is what makes it yours; the grade is only to show you what you had.

How sure?