Eric Bigelow · "The Cognitive Revolution" · How Agents Decide: Goodfire's Eric Bigelow on Critical Tokens, Phase Shifts, & In-Context Learning
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
Argues that most short-term gains will come from context engineering rather than building custom fine-tuned models per user
“Every SME should eventually fine-tune, post-train their own model if they can afford to, build one from the ground up, especially if they're in a very narrow domain. It's not that expensive to build a small language that caters to the needs of your customers.”
“You'll start by offsetting the short-term gains of the short-term losses, and then the long-term gains of the long-term losses”
“Portsmouth businesses should optimize for every AI model rather than backing one favorite. Your customers do not all use the same tool, so picking one just means going missing for everybody else.”
“You may need a bit of fine-tuning of your probes so that they work on each person's particular fine-tuned model”
The strongest new moments on your topics, in the speakers' own words. Free for three topics.
No card. No password. One-click unsubscribe.
Pro watches up to 25 names or phrases and emails you the morning after they are said.
See Pro, $29 a monthVerbatim quotes a short extract and links to the episode; the episode belongs to its publisher. Remove a show.