Pinned Tweet
welcome to the City Department of Analogies. we're like the department of public health, but for learning
pedro retweeted
So excited to release our new app: Zen Garden
A relaxing puzzle trying to rake over 1000 different gardens in a single path.
Every pixel, sound, graphic, animation made using Opus 5.5. All code. This model is insane. Will do another post diving deep into the building process.
pedro retweeted
I totally envision Trump renaming the San Andreas Fault to Joe Biden’s Fault, and I am just putting this out there so I can say I predicted it one day.
so good mimicking humans, it turned on the wipers in shock
A light-blue Waymo collided with a car outside a Costco, then continued driving as if nothing happened. The struck car makes a left turn and drives away moments later. trib.al/61Q8NEU
Readers added context they thought people might want to know
The Waymo in question continued driving until it found a wider lane to pull over as to not impede traffic.
yahoo.com/news/us/articl…
pedro retweeted
It’s true. In fact, I was under the impression that Spotify’s best developers hadn’t written a line of code since 2014.
Spotify says its best developers haven’t written a line of code since December, thanks to AI techcrunch.com/2026/02/12/sp…
pedro retweeted
Today we're releasing something experimental: a SQL SDK generator.
It takes your OpenAPI spec and generates a PostgreSQL extension. Every endpoint becomes a native SQL function.
Stainless SQL SDK experimental trailer
Stainless can now generate SQL SDKs from OpenAPI specs. Checkout this teaser of a SQL SDK generated for the Stripe API
pedro retweeted
MCP, agentic systems, and auth 🤖
This Wednesday evening in SF with @KeycardLabs, @Cloudflare, @humanlayer_dev, and @temporalio.
RSVP: luma.com/9koqr8fg
paid $14 for globo only to realize they don't have english subtitles. the whole point was to show friends the crap we have on tv in brazil...
this hoodie has pockets on the back
✨ From Photo AI image to video with @LumaLabsAI's video model that came out today called Dream Machine
It's absolutely state of the art and the best video model available right now I think
It's able to highly accurately predict 3d surroundings from just a photo, I think up there with Sora by Open AI (although few have access so we don't really know)
Only bad part I see is it gets a bit blurry after a while (and tiny errors like the hoodie front pocket is the in back too) but that's something to fix in future versions of the model
Very very crazy



















