Our AI agents started filing feature requests for their own tools.
In the days after I added the feedback loop, they submitted 50 requests. Each one captured what the agent wanted to do at the moment it got stuck.
Lo nuevo en https://t.co/VJYn364we6 : si entras y no hay nadie con quien competir, te matchea con una leyenda. Las leyendas son personajes conocidos por farmear aura.
Cada leyenda tiene diferentes chances de ganar. ¿Crees que podes ganarles?
Hice un jueguito: https://t.co/Ut0hFTHgiE
Competís contra otros farmeando aura. Otros jugadores hacen de jurado y votan por quién tiene más aura de los dos. Ganás, sumás aura, y la mostrás al mundo con tu perfil.
¿Sos la persona con más aura del planeta? Entrá y demostralo.
Si ganás = +1 aura. Si perdés = no perdés nada. Tu aura solo sube.
Sos anónimo: el rival solo ve tu handle, y recién después del duelo.
Cada partida queda grabada durante 7 días por si hay que revisar un reporte. Los bans son reales.
English spelling is rough. Anyone who has taught someone to read knows how many unusual rules there are, and how many exceptions to those rules exist. Just when a kid is starting to understand that there is logic to letters and sounds, the weird rules confuse them.🧵
Handwriting might not be that useful of an output, but as an input into learning to read, it's valuable. Dojo Sparks takes the same progressive approach to teaching writing letters. 🧵
How do you get a kid from sounding out s… a… t to reading "sat"?
In Sparks we build a ramp: three levels of scaffolding that fade as the kid's confidence grows. The same word the whole way through, each step building on the last, until you get the full word. 🧵
Blending is the first major cognitive leap for kids learning to read. It's a huge hurdle. We scaffold the difficulty with visuals that show two distinct concepts coming together. We call it our "Separate Together" activity. 👇
We are shipping 4 drawing modes for letter practice in @ClassDojo Sparks. Why? Because handwriting is crucial for reading development. It activates motor memory that helps kids internalize letter shapes. Here's how we're making it progressively challenging:
@jpsaraceno ¿cuánto oxígeno le quita realmente un auto, una combustión del auto, equivale a cuántas árboles se necesitan para evitar, justamente, que el árbol que da oxígeno. Indudablemente uno en la montaña, cuando ve realmente que no hay ningun árbol es que no hay oxígeno…..
What I find endlessly fascinating:
Some engineers really can't seem to grasp that LLMs are non-deterministic and how to build software taking that into account.
For others it immediately clicked, but for some it seems like there's a real mental barrier to accept it.