I've been having such an amazing time with Claude Code I wanted you to be able to have my *exact* skill setup: Introducing gstack, which you can install just by pasting a short piece of text into your Claude code

We collaborated with @NVIDIA to teach you about Reinforcement Learning and RL environments. Learn: • Why RL environments matter + how to build them • When RL is better than SFT • GRPO and RL best practices • How verifiable rewards and RLVR work Blog: unsloth.ai/blog/rl-enviro…

Simply adding Gaussian noise to LLMs (one step—no iterations, no learning rate, no gradients) and ensembling them can achieve performance comparable to or even better than standard GRPO/PPO on math reasoning, coding, writing, and chemistry tasks. We call this algorithm RandOpt.

Atoms. atoms.co/vision

That one neuron connects to about 7,000 others. Your brain has 86 billion of them. Do the math and you get somewhere around 100 trillion connections inside your head. More connections than stars in 1,500 galaxies. And each connection point is way more complicated than anyone

My conversation with Marc Andreessen (Marc Andreessen 🇺🇸), co-founder of a16z and Netscape. 0:00 Caffeine Heart Scare 0:56 Zero Introspection Mindset 3:24 Psychedelics and Founders 4:54 Motivation Beyond Happiness 7:18 Tech as Progress Engine 10:27 Founders Versus Managers 20:01 HP Intel

AI agents are getting deeper LangChain created DeepAgents, an open-source framework for building AI agents that can plan tasks, spawn sub-agents, manage files & maintain context Basically turning LLMs into autonomous systems for complex workflows GitHub:

Pull historical price data in 1 API call. Daily OHLC, volume, and adjustments. 14,000+ stocks. 20+ years.

Karpathy asked. I delivered. Introducing OpenSquirrel! Written in pure rust with GPUI (same as zed) but with agents as central unit rather than files. Supports Claude Code, Codex, Opencode, and Cursor (cli). This really forced me to think up the UI/UX from first principles

This paper is almost too good that I didn't want to share it Ignore the OpenClaw clickbait, OPD + RL on real agentic tasks with significant results is very exciting, and moves us away from needing verifiable rewards Authors: Yinjie Wang Xuyang Chen, Xialong Jin, Mengdi Wang

I am very happy that my survey paper, "Deep Learning for Solving Economic Models," is forthcoming in the Journal of Economic Literature (pending final replication checks, which should be quick). The paper benefited greatly from the editor, David Romer, five referees, and many

x.com/i/article/2033…

Just saw this GitHub project 🛡️ OpenViking is skyrocketing 📈. This could be the best memory manager for OpenClaw🦞! 👀 ✅ OpenViking (volcengine/OpenViking) is an open-source project released by ByteDance’s cloud division, Volcengine. It's exploding in popularity and could

Today, at the DARPA expMath kickoff, we launched 𝗢𝗽𝗲𝗻𝗚𝗮𝘂𝘀𝘀, an open source and state of the art autoformalization agent harness for developers and practitioners to accelerate progress at the frontier. It is stronger, faster, and more cost-efficient than off-the-shelf

Bolsonaro was loyal to Trump. Where is Trump now?

America is allowing Jair Bolsonaro to die. He’s a political prisoner. The Trump administration lifted tariffs on Brazil without demanding Bolsonaro’s release. Bolsonaro is a clear cut example of what Democrats wanted to do to Donald Trump. We cannot allow this.

Caught up with Andrej Karpathy for a new No Priors: on the phase shift in engineering, AI psychosis, claws, AutoResearch, the opportunity for a SETI-at-Home like movement in AI, the model landscape, and second order effects 02:55 - What Capability Limits Remain? 06:15 - What

Qwen3.5 27B is awesome (the entire family above 9B is impressive). You can now try it directly in your browser at SOTA speeds with whatever GPU you have: hf.co/spaces/Ex0bit/… My previous research in practice - The `Intel/Qwen3.5-27B-int4-AutoRound` is particularly good.