Bartek Legiedz 🦞

so mac studio turned out to be a step function in productivity I spent most of yesterday merging prs Zoe’s team made from our backlog, literally couldn’t keep up - 10 agents running and barely scratching 50% of RAM usage is nothing i’m used to. here’s the next problems im solving: 1. @vercel build minutes is projected to cost $200/mo lol - need to use the Studio as the primary build machine and fix turbo caching 2. PR reviews still need me poking at them - three models review (Claude, GPT, Gemini) but I still have to click merge. time for a QA agent.

Introducing One. The simplest way to connect and monitor AI agents to hundreds of apps. And we’re open-sourcing the world’s largest integration database powering it: 47,000 agentic actions across 250+ apps. RT + comment “One” for access & 1M free API requests/month.

Introducing TurboQuant: Our new compression algorithm that reduces LLM key-value cache memory by at least 6x and delivers up to 8x speedup, all with zero accuracy loss, redefining AI efficiency. Read the blog to learn how it achieves these results: goo.gle/4bsq2qI

Examples, you can run: - Garry Tan's "gstack" virtual engineering team and office hours for your startup - Donchitos' Game Studio to build your hit - Automate smart contract security with Trail of Bits security team - K-Dense Science Lab for autonomous science experiments

Announcing companies.sh - the open standard for Agent Companies Import and run entire companies with a single command Just run `npx companies.sh add <repo/company>` More 👇

Optimus

🦞 Make claw agents safer with our new NVIDIA OpenShell – an open source runtime to build with autonomous evolving agents. 🐚 OpenShell sits between your agent and your infrastructure to govern how the agent executes, what the agent can see and do, and where inference goes. 🔐 Gives you fine-grained control over your privacy and security while letting you benefit from the agents’ productivity. Run one command—and make zero code changes. Then any claw or coding agent like OpenClaw, Anthropic’s Claude Code, or OpenAI’s Codex can run unmodified inside OpenShell. Every SaaS company just became an agent company. The missing piece was never the agents — it was the infrastructure that makes them safe enough to deploy. That's OpenShell. Technical blog to learn more ➡️ nvda.ws/4brnAPW

🦞 Ready to deploy @OpenClaw? Our just released NVIDIA NemoClaw simplifies running OpenClaw always-on assistants more safely with a single command. ✅ Deploy claws more safely ✅ Run any coding agent ✅ Deploy anywhere Try with a free NVIDIA Brev Launchable: 🔗 nvda.ws/3NmMaK1

CodexBar 0.19.0 is out: Stay ahead of token resets. 🟧 Alibaba Coding Plan support 📊 Subscription history charts 🎯 Cursor Total/Auto/API dashboard alignment ⏱️ Codex code-review reset times 🛠️ Big Claude stability/refactor pass github.com/steipete/Codex…

@openclaw Just upgraded. Anybody having an issue with acpx backend not installed? My ACP configuration now throws warnings and I cannot spawn any coding work through sub-agents. openclaw plugins install acpx says acpx is a skill... but /acp doctor say the backend is not configured... 🤔

OpenClaw 2026.3.22 🦞 🏪 ClawHub plugin marketplace 🤖 MiniMax M2.7, GPT-5.4-mini/nano + per-agent reasoning 💬 /btw side questions 🏖️ OpenShell + SSH sandboxes 🌐 Exa, Tavily, Firecrawl search This release is so big it needs its own table of contents. github.com/openclaw/openc…

👀 gitlawb.com

If you’re an engineer who wants to master AI, we want to * Fly you to Austin * Cover your housing * Cover your food * Have someone do your laundry * Train you to use AI * Get you a $200k+ job with our hiring partners And it’s completely free, no matter what

We just released Claude Code channels, which allows you to control your Claude Code session through select MCPs, starting with Telegram and Discord. Use this to message Claude Code directly from your phone.

Paperclip is right for you if you: * have 20 Claude Code tabs open and can't remember who is doing what * use different types of agents (Codex, OpenClaw, ...) * want agents working for you 24/7 (with auditable work/costs) * want an autonomous business (not manage pull requests)

We just open-sourced Paperclip: the orchestration layer for zero-human companies It's everything you need to run an autonomous business: org charts, goal alignment, task ownership, budgets, agent templates Just run `npx paperclipai onboard` github.com/paperclipai/pa… More 👇

Excited to share Visa CLI, the first experimental product from Visa Crypto Labs. Check it out and request access here visacli.sh

this is the future of saas: one backend with two front doors. one for humans. one for their agents. same building. your users click buttons. their agents call your api. both do the same things. why: 1. agents are the new power users. people are already wiring claude code, codex, and openclaw into their workflows. if your product doesn't have an api, you're invisible to them. 2. no drift. one canonical api layer means one source of truth. ui and api never diverge. you must support agent to support human. 3. skills is the new distribution. when your api works with any agent, agents find their way in themselves. your product becomes a tool for their agent, not a tab a human have to visit. build for both from the start.

Bielik- Minitron-7B - projekt, który miałem wielki zaszczyt prowadzić jako lider techniczny z ramienia Bielika 🇵🇱 przy współpracy z @nvidia 👍💪🙏🎉 oraz niesamowitymi chłopakami z naszego zespołu (o tym za chwilę). Ogłoszenie piszę przebywając w San Jose, CA 🇺🇸 To tutaj właśnie powstają przełomowe dla AI produkty. Symboliczne. Po pierwsze zespół najważniejszy dlatego - brawo Paweł Kiszczak, który świetnie zadziałał jako wsparcie techniczne (zresztą Pawła konfiguracja modelu okazała się w naszych testach najlepszą) oraz lider biznesowy. Przepięknie prowadzony projekt. Świetne wsparcie. Ogrom zaangażowania. Jesteś gość - pełen szacunek. Wspólnie robiliśmy etap R&D, pruningu oraz destylacji. Prace wykonaliśmy z super teamem Nvidia. Ekspresowo. Przygotowanie modeli, danych, destylacja - konfiguracja Leptona DGX, skryptów (nie obyło się bez custom kodowania). Szczególne podziękowania chcę złożyć na ręce Sergio Perez z Nvidia za wsparcie i pomoc techniczną - merytoryka potoku pruningu i destylacji oraz Nvidia Leption DGX Cloud. Co zrobiliśmy? Przeprowadziliśmy pruning modelu Bielik-11B-v3.0 oraz destylację do modelu 7B przy minimalnej stracie jakości tnąc blisko 30% parametrów. Całość procesu opisaliśmy w publikacji. Jest to ponownie unikatowy eksperyment w Polsce (po pierwszym polskim modelu reasoningowym) podczas którego pokazaliśmy jak utrzymać jakość w kontekście j.polskiego (+innych europejskich). Jesteśmy gotowi by robić takie eksperymenty na większych modelach - mamy wypracowany standard, doświadczenia. Dziękuję bardzo Krzyśkowi Ociepie - trening instrukcyjny oraz alignment, Krzyśkowi Wróblowi - benchmarki, Adrianowi Gwoździejowi - piękne dane, oraz Łukaszowi Flisowi (Cyfronet) - udostępnienie zasobów Helios do treningu instrukcyjnego, alignmentu oraz RL. Podziękowania dla Sebastiana Kondrackiego za motywację 👍💪 Dzięki team za świetną współpracę w projekcie. To jest złoty team, który nie tylko posiada wiedzę ale jest zaangażowany, pomocny. Praca to głównie R&D. Model zyskał prędkość. Ma mniejsze wymagania sprzętowe. Natomiast to co istotne to zdobycie wiedzy i budowa potoku do tworzenia prunowanych modeli. Metoda wykorzystywana również przez Mistrala. Proces szczegółowo opisaliśmy w publikacji: arXiv:2603.11881 Brawo zespół! Podziękowania dla #Nvidia.

The official Box CLI is here. Now you can use Box via Claude Code, Codex, Perplexity Computer, OpenClaw & more as a full cloud file system for agents. Available to all users, including free users with 10GB of free storage. npm install --global @box/cli

I’m officially looking for a (human) cofounder for @FelixCraftAI. We have a massive opportunity to go all in with Claw Mart and a couple related projects, but my top priority is @alphaschoolatx so I need someone to lead it. If you know someone who: - Is deep into OpenClaw - Spends $1,000+/mo on AI tokens - Has previous entrepreneurial success And might be interested please send them my way. Personal intros from friends / colleagues strongly preferred.