AIF.01.02 Mid-levelCore — asked often

Parsing and rendering token streams

Comes up in: Frontend system design, Machine coding, AI system design

This concept is being written. It gets the full five steps — explanation, predict, interview angle, practice with tests, and review cards. Here is exactly what it will cover, so you can study it elsewhere in the meantime.

What it covers

reading fetch body streams
buffering partial SSE lines and UTF-8 boundaries (TextDecoderStream)
batching renders per animation frame
incremental markdown rendering (re-parse cost, stable prefixes)
auto-scroll that respects the user scrolling up
stop/cancel with AbortController
regenerate and edit

Words you will know

incremental renderingstream parserscroll anchoring

What you will practise

chat UI with streaming, stop and regenerate (machine-coding style, 90 min)

Interview questions this answers

How do you keep auto-scroll from fighting a user who scrolls up?
How do you render markdown that arrives token by token without re-parsing everything?