chrbx128 (@chrbx128) 's Twitter Profile
chrbx128

@chrbx128

🇺🇦 software engineer. current interests: AI, psychology, distributed systems and self-efficiency.

ID: 827126347948519424

calendar_today02-02-2017 12:07:43

588 Tweet

20 Followers

246 Following

Mark Kretschmann (@mark_k) 's Twitter Profile Photo

OpenAI just released Prompt Packs for every job. Using these prompts, ChatGPT by OpenAI can do any job like a human would: * Engineering * Manager * Product Developer * Marketing * Artist * Plumber ...and much more!

OpenAI just released Prompt Packs for every job. Using these prompts, ChatGPT by <a href="/OpenAI/">OpenAI</a> can do any job like a human would:

* Engineering
* Manager
* Product Developer
* Marketing
* Artist
* Plumber

...and much more!
DAIR.AI (@dair_ai) 's Twitter Profile Photo

This paper is worth reading carefully. It introduces System 3 for AI Agents. The default approach to LLM agents today relies on System 1 for fast perception and System 2 for deliberate reasoning. But they remain static after deployment. No self-improvement. No identity

This paper is worth reading carefully.

It introduces System 3 for AI Agents.

The default approach to LLM agents today relies on System 1 for fast perception and System 2 for deliberate reasoning.

But they remain static after deployment. No self-improvement. No identity
Akshay 🚀 (@akshay_pachaar) 's Twitter Profile Photo

This is the DeepSeek moment for Voice AI. Chatterbox Turbo is an MIT-licensed voice model that beats ElevenLabs Turbo & Cartesia Sonic 3! - <150ms time-to-first-sound - Voice cloning from just 5-second audio - Paralinguistic tags for real human expression 100% open-source.

Chubby♨️ (@kimmonismus) 's Twitter Profile Photo

Tencent releases WeDLM-8B-Instruct A diffusion language model that runs 3-6Ă— faster than vLLM-optimized Qwen3-8B on math reasoning tasks. -3-6Ă— faster than vLLM-optimized Qwen3-8B on math reasoning tasks - Outperforms base Qwen3-8B-Instruct on most benchmarks - Native KV

Tencent releases WeDLM-8B-Instruct

A diffusion language model that runs 3-6Ă— faster than vLLM-optimized Qwen3-8B on math reasoning tasks.

-3-6Ă— faster than vLLM-optimized Qwen3-8B on math reasoning tasks

- Outperforms base Qwen3-8B-Instruct on most benchmarks

- Native KV
Robert Youssef (@rryssf_) 's Twitter Profile Photo

This paper from Stanford University tackles a question that’s been debated for decades in quantitative finance: can raw market data alone meaningfully predict stock price direction, without handcrafted indicators or heavy financial feature engineering? The author proposes a

This paper from Stanford University tackles a question that’s been debated for decades in quantitative finance: can raw market data alone meaningfully predict stock price direction, without handcrafted indicators or heavy financial feature engineering?

The author proposes a
alphaXiv (@askalphaxiv) 's Twitter Profile Photo

RoPE is fundamentally flawed. This paper shows that RoPE mixes up “what” a token is with “where” it is, so the model can’t reliably reason about relative positions independently of token identity. Eg. the effective notion of “3 tokens to the left” subtly depends on which

RoPE is fundamentally flawed.

This paper shows that RoPE mixes up “what” a token is with “where” it is, so the model can’t reliably reason about relative positions independently of token identity.

Eg. the effective notion of “3 tokens to the left” subtly depends on which
🍓🍓🍓 (@iruletheworldmo) 's Twitter Profile Photo

new paper worth reading if you want to understand what’s actually happening inside these systems. researchers showed that standard transformer training implicitly runs an EM algorithm. attention = E-step (soft assignment of which positions matter). values = M-step (updating

new paper worth reading if you want to understand what’s actually happening inside these systems.

researchers showed that standard transformer training implicitly runs an EM algorithm. attention = E-step (soft assignment of which positions matter). values = M-step (updating
KNOX (@knoxtwts) 's Twitter Profile Photo

claude, my AI overlord, i need you to scrape every $1m+ info launch from 2025, reverse engineer the copy, steal the funnel structure, build me a course on whop, price it at $297 with a fake discount from $997, set up financing, reverse engineer every viral info tweet from january

Justin Farrugia (@justinmfarrugia) 's Twitter Profile Photo

honestly insane how fast everything is moving it was barely a week ago when I was asking how to get good design outputs with Claude Code now we have an entire stack for agentic UI

Zac (@perceptualpeak) 's Twitter Profile Photo

holy shit it fucking WORKS. SMART FORKING. My mind is genuinely blown. I HIGHLY RECCOMEND every Claude Code user implement this into their own workflows. Do you have a feature you want to implement in an existing project without re-explaining things? As we all know, the more

holy shit it fucking WORKS. 

SMART FORKING. My mind is genuinely blown. I HIGHLY RECCOMEND every Claude Code user implement this into their own workflows. 

Do you have a feature you want to implement in an existing project without re-explaining things? As we all know, the more
nader dabit (@dabit3) 's Twitter Profile Photo

Someone created Awesome Ralph and there is a lot of good stuff here for learning and building agent loops github.com/snwfdhmp/aweso…