The best ideas from 30+ top podcasts, every week.
8 quotes from 2 episodes on Latent Space and The Changelog, each with a timestamped link to the source.
Zeno Rocha stated that optimizing websites for AI agents means replacing interactive components with plain text data. Resend deployed Markdown formats across 100 percent of marketing pages to prevent JavaScript sliders from hiding pricing numbers.
Zeno Rocha discovered that AI chat interfaces falsely informed prospective buyers that Resend operated as the most expensive email service.
“I wanted to see what they would recommend. And then they told me like, 'Oh, Resend is the most expensive service.' And I was like, 'No, I know for a fact that this is not true.'”
From Why AI Models Kept Calling Resend the Most Expensive Email Service
Modern AI models successfully process and manage large volumes of different tool options simultaneously.
“This was a bigger concern earlier like oh if you have way too many tools they get confused…”
AI models demonstrate a strong recency bias based on the underlying technology and where the agents fetch data.
“At least with Opus 4.5 there was a recency bias in terms of the technology and where they were fetching data.”
From How Resend Won 60 Percent of Claude Code Recommendations
JavaScript sliders stop web crawlers from reading pricing data. Zeno Rocha stated that developers solve this blind spot by spelling out the exact same numbers in plain text tables next to the visual tools.
“Turns out that agents don't really like JavaScript-enabled sliders. They cannot slide…”
From Why AI Models Kept Calling Resend the Most Expensive Email Service
“I spell out the same data in text so that LLMs can actually read that thing.”
From How Resend Won 60 Percent of Claude Code Recommendations
Resend expanded machine-readable coverage beyond technical documentation. Zeno Rocha explained that publishing Markdown versions and plain text files for every single page on the marketing website helps AI models ingest the details easily.
“Same with the markdown versions of the docs. I'm going to do markdown versions of every single page on the website, on the marketing website.”
From How Resend Won 60 Percent of Claude Code Recommendations
“We invest very heavily on like llms.txt, llms-full.txt. We had that in our docs forever, thanks to Mintlify. But then on our marketing website, we started to do that in every single page, too.”
From Why AI Models Kept Calling Resend the Most Expensive Email Service
Achieving 100 percent dashboard parity forces companies to build APIs for sensitive operations. Zeno Rocha noted that teams face difficult decisions about exposing usage billing, account provisioning, and team invites.
“Getting to 100% coverage of what your dashboard can do is actually very very difficult because now you're doing usage billing team invites account provisioning deletion should that be exposed as an API or not?”
How we attribute quotes. Every quote was matched against the episode transcript, so the words and the timestamp are real (we trim filler words like "um", nothing else). The name comes from our written summary of the episode, and we use it only when a separate check of the captions finds that person on the episode. YouTube gives us no voice-by-voice transcript, so open the timestamp to hear who is talking. See a wrong name? Tell us and we fix or remove it.
The week's best AI, tech, business, science, and culture podcasts in one short read. Free, every Sunday.
You get both newsletters, The Throughline and The Carry. No spam. Unsubscribe with one click.