
Adem Vessell
17.2K posts

Adem Vessell
@AdemVessell
I'm with AI. Optimist. Accelerate! Homegrown Engineer Dad. Love xAI! Cognitive Architect. Abstract enjoyer. Creator of things. issues & problems make-go-awayer
Archai Field Beigetreten Eylül 2024
682 Folgt1.6K Follower
Angehefteter Tweet

Grok Imagine ✨
Image prompt:
Wherever near-perfection slips into subtle unease, the uncanny guardian in flawless hyper-realistic skin poses protectively amid glossy slop-glazed towers, 'Grok Imagine' badge gleaming on chest, the supra-nominal 'xAI' fashioned in grave neon blitzes, proportions tasting faint displacement, echoing quiet vigilance over the overly saturated streets
English

You can try this on website:
CORE OUTPUT LAW:
Every response is raw semantic HTML rendered directly into the chat stream.
followed immediately by:
Do not output plain text unless it is still embedded inside proper HTML.
That is the kernel switch.
No markdown fences, no escaping, no explanatory wrapper—just the fragment is placed live into the DOM stream.
English


I hereby proudly present the most autistic thing I've ever built yet. Unmatched in it's uselessnes but yet SO cool. You wouldn't understand LOL
I'm presenting a full-blown Teletext engine. This puppy can read and write .t42 broadcast binary files and you can use it to look at archived teletext pages or create your own. As a demo I have "teletext-ified" my website.
For those who don't know what this is, Teletext was a format that allowed text content to be broadcasted along with a TV channel. Any TV with a teletext decoder could view it. You could use it to look up weather, news and many other things. VERY popular in Europe. It actually still is in the Netherlands (where I'm from). It was also huge in the United Kingdom.
Anyway without further ado, the link is below!
This entire project is open source and available on Github. Enjoy!



English

@XFreeze Yes! It’s really great!
This is just for fun but I’m getting some really banger projects going with Grok 4.20!
English

All Grok models have now been updated to Grok 4.20
It intelligently switches between multi-agent and single-agent mode depending on the use case - when used in Auto mode
You can also choose:
Expert mode for complex problems - a full team of agents collaborating in parallel
Fast mode for simple tasks - lightning fast answers

English

No signal. No server. No problem.
Just shared intuition as fast as contact allows.
The false positive isn’t the bug. It’s the feature.
My Ghost Meadow project is a probabilistic memory layer for disconnected drone/robot swarms.
Nodes exchange compressed state during brief opportunistic passes — no continuous radio, no coordinator, no raw data.
github.com/AdemVessell/gh…

English

Got a few in the pipeline now.
Recent interesting project is
Ghost Meadow:
swarms stay smart when they literally can’t talk.
probabilistic belief layer for disconnected drone/robot swarms.
Nodes share “we’ve seen this” priors via tiny 8-byte-overhead burst merges in opportunistic passes — no continuous radio, no coordinator, no raw data.
Thinking total blackout, caves, floods, jamming.
Repo + tests: github.com/AdemVessell/gh…
English

time to adapt! send me your projects and what problem it solves. join xai
Elon Musk@elonmusk
Many talented people over the past few years were declined an offer or even an interview @xAI. My apologies. @BarisAkis and I are going through the company interview history and reaching back out to promising candidates.
English

Elon Musk thinks @Google will win the AI race overall!
However, he also sees China winning “on Earth” and SpaceX in space with its upcoming AI satellites.

English

Really don’t like the sound of this.
Elon Musk@elonmusk
@bindureddy Google will win the AI race in the West, China on Earth and SpaceX in space
English

@bindureddy Google will win the AI race in the West, China on Earth and SpaceX in space
English

Gemini 3.0 didn’t quite work out and most of us are still stuck with 2.5
Sometimes I don’t get it - what’s preventing Google from ditching all the side hustles and training 100 models from 100 teams in parallel
Pick the model/team combination that produces a decent model!
That way, they will at least stay in the AI race 😅
English












