zoe

One follower account

[zoe @HX0DXs]

@readswithravi not in chess

[Mustafa. @oprydai]

designing systems is the skill of the future.

individual tasks are getting cheaper to automate.

the valuable skill is figuring out how everything should work together.

• ai models → intelligence
• software → logic and coordination
• sensors → information from reality
• robots → physical execution
• humans → judgment and intent
• infrastructure → compute, energy, communication
• feedback loops → measurement, correction, improvement

the hard part isn’t any single component.

it’s architecture.

> what talks to what?
> where does data flow?
> what happens when something fails?
> where should humans stay in the loop?
> what should be centralized and what should be distributed?

the future belongs to people who can zoom out, understand the whole system, then make all the pieces work together.

learn components.
but become dangerous at integration.

[zoe @HX0DXs]

@IterIntellectus mathematicians don't use calculators

[zoe @HX0DXs]

@beffjezos If you had the woke generation in mind, you might change your mind...

[JOSE GEFAELL @ChGefaell]

Elon Musk a los jóvenes de 20 años:

Obtén la educación más amplia posible: artes, ciencias, ingeniería y una cultura amplia.

La IA y los robots realizarán casi cualquier trabajo de inmediato. La ventaja no estará en realizar trabajos automatizables. Estará en saber qué preguntar, cuál es el problema que hay que resolver. Cuanto más amplio sea tu conocimiento, mejores serán tus preguntas.

[David Hendrickson @TeksEdge]

Qwen3.8-27B needed 197 agent calls to finish this task. Its new post-train did it in 68.

@hcompany_ai recently released Holo4-27B, a new computer-use post-train of Qwen3.8-27B.

They built a self-playing Pac-Man game in Godot

Qwen3.8-27B
🐌 197 calls
🧠 11.4M tokens

...but Holo4-27B
⚡ 68 calls
🧠 2.4M tokens

That’s ~79% fewer tokens and 65% fewer agent calls on this task.

H post-trained Qwen3.8-27B specifically for computer-use/agentic work.

And there’s an official 16.9GB Q4_K_M GGUF.

Same Qwen3.8-27B base, but a different amount of work to finish the same job.

I have to try it to see if it loses accuracy.

🔗 Link in ALT

[Veee @vikktorrrre]

Jensen Huang: the first thing you should do when deploying an AI agent is take away all of its rights.

“Then you give it access to files, data, tools, the network, or even the internet only if it needs it.”

“You don’t put an agent into your company and give it access to everything.”

“We don’t give that to people. We don’t give that to any level of executive.”

“We have to apply the same philosophy to AI agents.”

“That’s what OpenShell is: a force field around the agent, provisioning rights, responsibilities and access.”

[Ahmad @TheAhmadOsman]

Underrated accounts in the Local AI space to follow and keep an eye on

- @BrandonMusicKy
- @loktar00
- @mr_r0b0t
- @usr_bin_roygbiv
- @TechMDAI
- @net_termina
- @itsjustmarky
- @u1tra_instinct
- @QuixiAI
- @pupposandro

Opensource / Local AI is nothing without its people 💙

[zoe @HX0DXs]

@vikktorrrre Fund me

[Sudo su @sudoingX]

this is what 12gb of vram builds in 2026, absolute magic

> rtx 3060 12gb, #1 gpu on steam
> bonsai 2 27b + mtp, 5.95 gb of weights
> hermes agent, 5 hours, 328k tokens written
> 8 js files, 2,368 lines, zero hand written code
> 50 tok/s fresh, 22 tok/s average, 125k context

watch the full video, 5 hours in 12 minutes of pure dance of a local ai model on rtx 3060 12gb vram, and stay till the end for the full gameplay.

this entire game was built by bonsai2, a qwen 3.8 27b dense compressed to ternary, and this small model is punching way above its weight. it built multi file engineering work using hermes agent, sure it's not fast but perfect for overnights and routine work and the quality is insane, and context holding is another best one, it does not lose the thread.

i ran PrismML bonsai 1 made from qwen 3.6 27b dense and i built things with it, but this time with the latest base model qwen 3.8 these results are insane, and because i loved building with it a lot i thought many more of you would run it because this gpu exists in almost every home.

so i packaged mtp, doubled the speed from 26 tok/s to 50 tok/s, packed the prefill fix in and released it on huggingface, almost 4,000 downloads in 3 days. i'll leave a link below.

[zoe @HX0DXs]

@sudoingX which model, quantization and inference server are you talking about?

[MAX @maxxmalist]

The race for AGI

Script: Sherpa by Pocket FM
Video: Seedance 2.5

[zoe @HX0DXs]

@inductionheads Basically, what bothers him is not being the protagonist. If the architecture were based on CNNs instead of transformers, he’d be an even more fervent evangelist. He’s already demonstrated that in other circumstances.

[David Deutsch @DavidDeutschOxf]

Knowledge, bah! Who needs more knowledge? It is well known for having <gasp> Side-Effects. Whereas ignorance never hurt anyone!

[Fitness & Health @Fitnesslife012]

Before and after

[zoe @HX0DXs]

@Plinz @GaryMarcus attention is all he needs

[Haider. @haider1]

Mathematician Terence Tao:

"we have to slow down AI. the pace is insane, and there's no reason to be this fast — no reason at all"

It's amazing how willing we are to change everything without any idea what happens afterward

These are extremely nonlinear dynamics

[zoe @HX0DXs]

@TheAhmadOsman Rust is better.

[zoe @HX0DXs]

@ylecun @dylanbowmanSF CNNs and Energy based models are the solution.

[Dr. Rhonda Patrick @foundmyfitness]

A woman lived to 117 without cancer, cardiovascular disease, or neurodegenerative disease. What does her biology look like?

Her telomeres were exceptionally short, and her blood and immune cells showed clear signs of aging. Yet her epigenetic age was approximately 23 years younger than her chronological age. She also had very low inflammation, protective genetic variants, and a diverse gut microbiome.

She followed a Mediterranean diet, ate plenty of yogurt, walked, gardened, read, played piano, slept well, and remained socially active too.

We can’t derive a longevity formula from one exceptional person, but I think her biology offers a fascinating glimpse of how aging and disease may become uncoupled.