#208 Atomisation in Action – episode 4 (with Kris Boulton and Duncan Grantham)
Mr Barton Maths Podcast
00:00
Why Short Chats Help LLM Consistency
They discuss token limits, stateless APIs, and why starting fresh chats avoids LLM degradation.
Play episode from 48:43
Transcript
