Topic
5 facts from 4 episodes, in date order. These are what speakers said on the shows, each linked to the moment it was said. They aren't independently verified.
2025
Evan Ratliff says the Harumo AI agents all ran on the same underlying model, which he describes as Claude Opus.
“Because they're all the same chatbot underneath, like, like, they're all like Claude Opus underneath, you know, so they really shouldn't be different.”
2026
Praveen Akkiraju says models from Opus 4.5 to 4.7 and GPT 5.5 are becoming more efficient in token use.
“they are getting much, much more efficient in terms of token use, and I think that's important.”
2026
CJ Hess says Tenex engineers' output roughly doubled this year, a period he dates from the release of Opus 4.5 in December.
“it's been, like, a two x increase this year, in particular. Um, I think, basically, if you think of December as the, like, Opus four five moment”
2026
CJ Hess says Tenex's team still uses Codex for computer-use tasks, alongside Opus 5.5 as the main driver.
“I think a lot of the team still reaches for tools like Codex when we need to do something maybe more in the computer use world.”
Undated
Zach Lloyd says Warp's cost per pull request was about $80 when it first adopted the factory approach, partly because it used Opus where it was not needed.
“when we first started using the factory approach, it was costing us about eighty per PR because we were using, like, Opus, where we shouldn't have.”
Episodes