Inference Engineering

English

Tiếng Việt

English

Tiếng Việt

Appearance

Inference Engineering NotesFrom one token to a production serving stack

Study notes for the fanout.sh Inference Engineering course. Every number re-checked, every figure redrawn.

Start with lesson 01
Revision packs

61 lessons

Speed physics, KV memory, continuous batching, PagedAttention, prefix caching, quantization and GPU profiling.

Mock-interview packs

Seven themed packs, each card a question with a 60-second answer, the key numbers and the usual trap.

Song ngữ

Switch to Tiếng Việt from the language menu. Lessons not yet translated show the English original.

About

Unofficial personal notes, not affiliated with or endorsed by fanout.sh. They summarize, not replace, the course. Source: S1LV3RJ1NX/inference-engineering-notes.