“‘One of the biggest news is GPT 5.6 wrote more efficient kernels. So Terra and Luna could be 80% cheaper.’”
♫
Said on Latent Space: The AI Engineer Podcast
Dave Slutzkin · The Startup Podcast · Applied Wizardry: Cutting AI coding token costs, with Dave Slutzkin
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
Assign specific model families to task types: use Astra for heavy thinking, Sonnet/Luna for sub-agents, avoid Terra and Haiku
“‘One of the biggest news is GPT 5.6 wrote more efficient kernels. So Terra and Luna could be 80% cheaper.’”
“Claude Haiku 5.5 is now at 21 cents, and GPT-6 Luna is at 7 cents, so a significant discount.”
“As they're answering things where I give them an assignment, automatically spawn a sub agent that traverses the vault for pertinent.”
“I avoid recurrent meetings… Slack and huddling Slack… for tasks, communication, chats”
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.