Angel Chen ▄ ▀

ideo-syncretic ✦ Generalist dev + design. Partner to @astralwave. Sensible ∩ friendly tech.

I love good stories and robots ✺

[Angel Chen ▄ ▀ @ideosyncretic]

I’d like to see a cultural counterpoint to Black Mirror, but today’s smart creative optimists don’t write fiction when they can work in tech directly.

Leaving anti-tech low-agency romantics to fill speculative pop culture with silly cynical dystopia porn or naive solarpunk.

[David @DavidSHolz]

we should all feel an obligation to tell stories about a human future that we actually want to be a part of.

[Angel Chen ▄ ▀ @ideosyncretic]

@willdepue Apparently it’s not playing in true live realtime though?

Just executing a pre-scripted sequence

[Angel Chen ▄ ▀ @ideosyncretic]

@januarycomputer @qibiz_me nah the long tail of small specialized models will co-exist alongside big general models, even as inference costs drop

cheaper, local, on-device, low latency etc

[Angel Chen ▄ ▀ @ideosyncretic]

I’m actually surprised this isn’t more common by now. It’s so intuitive to assign agents roles by specialism/domain expertise (plus a “Chief of Staff” main bot coordinator), and interact within a Slack-style chat interface. It’s such a familiar mental model for work.

You don’t heavily supervise or micromanage your coworker’s tasks, but you can dive deeper if needed. Mostly you just talk to them.

I’m so glad Grok Bot’s taken this direction, with great organic appeal and user adoption. Looking forward to seeing more good design from this team!

(Compare this to the utter crickets around the poorly-thought-out ChatGPT Work. Even Chat could benefit from domain/role-based “agent” division, imo)

[John Bai @johnbai]

Crazy that it’s only been 3 weeks since we launched Grok Bot.

Here’s a look at how we designed v1, and a glimpse of where we’re headed next.

https://t.co/Hgk1vQtTP7

[Angel Chen ▄ ▀ @ideosyncretic]

@seflless @pvncher @Dimillian Oh I hadn’t seen that one, very cool

[Angel Chen ▄ ▀ @ideosyncretic]

@seflless @pvncher @Dimillian There’s this!

[Peter Gostev @petergostev]

GPT-6-Astra: Peacock in SVG

[Angel Chen ▄ ▀ @ideosyncretic]

@stevegaan @johnbai @bot Now I kinda wanna see a whole conga line of them!

Least favorite is the squinty slanty eyes. I know grok’s got an edgy deviant brand but it looks a bit constipated instead of impish lol

[Angel Chen ▄ ▀ @ideosyncretic]

I think the marketing angle is off.

It’s great that this advances model understanding of UI.

But it makes very little sense as a final artifact or replacement for real UI, as the “everything is pixel/video gen” marketing suggests.

Non-deterministic UI is a horrible idea for most use cases shown in the demo.

[Angel Chen ▄ ▀ @ideosyncretic]

@Shedletsky @matthuang Nobody said each person only *uses* 1 GPU...

[Angel Chen ▄ ▀ @ideosyncretic]

@threepointone #PHASEONE10841DidNothingWrong

[Angel Chen ▄ ▀ @ideosyncretic]

@wessorh @matthuang If you truly prompt it to fix *only spelling errors* without changing your words, it’s unlikely to be flagged as AI

[Jon Stokes @jon_stokes]

https://t.co/UQ3KeMUhHD

[Surya @kickingkeys]

You can just RL a coding model to paint with javascript btw

[Angel Chen ▄ ▀ @ideosyncretic]

@khoomeik @kickingkeys’ post got me thinking about this: https://t.co/UQ4HOhrfvB

[Surya @kickingkeys]

You can just RL a coding model to paint with javascript btw

[Angel Chen ▄ ▀ @ideosyncretic]

@krishnanrohit Why not @thinkymachines

[Angel Chen ▄ ▀ @ideosyncretic]

@evilmathkid @chris_j_paxton Then you make ’em race dirty. Mad Max that shit with robot horses and chariots and trucks and bikes

[Angel Chen ▄ ▀ @ideosyncretic]

@adcock_brett Didn’t expect that fisheye lens. I suppose this gets corrected/compensated for

[Jordan Dworkin @jddwor]

In 1843, Elizabeth Barrett Browning wrote to a friend after seeing her first daguerreotype: “The very shadow of the person lying there fixed for ever! … I would rather have such a memorial of one I dearly loved, than the noblest artist’s work ever produced.”

Our lives and homes are filled with things that once inspired similar awe. I tried to collect those feelings, in quotes like the one above, from people for whom what is now ordinary was brand new. A friar who met the inventor of eyeglasses, a founding father on the new smallpox vaccine, a woman in Iowa on her family’s first indoor bathroom. To look at a modern apartment through their eyes: https://t.co/3Nsti34aS2

[Angel Chen ▄ ▀ @ideosyncretic]

This is the right answer. ChatGPT’s “Chat/Work” distinction is unnecessary and confusing.

My work should move fluidly along an agentic continuum:

discuss → plan → inspect → execute → review → revise → execute (again)

That’s how people naturally work.

It’s bad UX to make us decide upfront which “pocket” a nascent piece of work belongs in, then leave the conversation stuck there forever.

“Chat” already handles agentic tasks via apps & tools.

“Work” shares the limited Codex/agentic usage pool. Presumably because agentic execution is expensive.

If so, then just meter the expensive execution, rather than tying usage to an arbitrary Chat/Work distinction.

Is there some underlying architectural, safety, or operational constraint behind the split?

Why can’t the same conversation fluidly shift to the heavier “Work” agent runtime as needed, then back to the lighter “Chat” mode?

[joshpuckett @joshpuckett]

@thsottiaux Kill Chat / Work tab