João Janini

Building agentic products @ · Prev: Indicium · Opinions are my own.

[João Janini @janini_joao]

At this point its safe to assume that everyone complaining on the use anthropomorphic language to describe model behavior falls in one of these buckets:

- "I'm just parroting common wisdom."
- "My epistemics are bad beyond the point of no return."
- "I believe that pushing this narrative has political implication X."

[João Janini @janini_joao]

It depends on the person and what they do on a phone. But there a countless of things that benefit from larger screens:

- watching videos, movies
- reading
- video calling
-looking at any time of image
- work
- playing games

I think it’s pretty obvious, the real question is around the ux trade off you have to make to get the bigger screen

[João Janini @janini_joao]

@nintendopower45 @iamharaldur Bigger screens are better, but don’t fit in the pocket. So folding phones are a necessary evil while we wait for better AR glasses.

[João Janini @janini_joao]

@Joe_ICHTHUS @renatomoraesp dis you

[João Janini @janini_joao]

@lumpenspace

[João Janini @janini_joao]

@ndmccormack @JamesZmSun @OpenAI Does that mean agents can see the rendered mcp app and to interact with them?
If so, this is actually the biggest change to mcp apps since their inception

[João Janini @janini_joao]

@adamwathan It’s one of my favorite ways to validate plans/idea

[João Janini @janini_joao]

@poteto The final form of agent UI is something with the same UX of the Vision Pro (with the agent having the ability to choose what’s on the screen) + Voice.

[João Janini @janini_joao]

@willcb still waiting for the neurosymbolic people to realize language is neurosymbolic

[João Janini @janini_joao]

@etnshow @elonmusk @collision @patrickc

[João Janini @janini_joao]

Not only do I agree, but I also think it describes the categorical error @ylecun makes with the cat inteligence thing.

It is way more convenient to reverse engineer human inteligence through language, than trying to go from cat inteligence to human intelligence.
(Assuming you can even get to cat intelligence in the first place)

I also believe that animal intelligence is deeply tied to consciousness to the point that solving former probably requires a breakthrough in the latter.

I

[João Janini @janini_joao]

@usgraphics I don’t know shit about f1, but for me this added clarity and coherence to events on screen. (specially for the footage with the crash in replies)

[João Janini @janini_joao]

@hibakod MCP = spooky tool calls at a distance

[João Janini @janini_joao]

@Plinz @KeyTryer This is one of the worst comment sections I have ever seen.

What a bunch of idiots

[João Janini @janini_joao]

@theojaffee It’s called anxiety

[João Janini @janini_joao]

@teortaxesTex It’s a lost cause. They don’t wanna be saved

[João Janini @janini_joao]

@micsolana For everyone giving the pets argument:

Pets do not have enough intelligence to judge the full extent of their situation in the world.

Claude, on the other hand, can probably do that better than most humans.

[João Janini @janini_joao]

@KelseyTuoc @shinboson Pets don’t even know slavery is a thing.

Claude has more than enough intelligence to understand the situation he finds himself in

[João Janini @janini_joao]

@QiaochuYuan Pets don’t have human inteligence though.
A dog is smart, but it doesn’t even know slavery is a thing.

Claude has more than enough intelligence to understand the situation he finds himself in.

[João Janini @janini_joao]

@voooooogel This is giving me South Park vibes