Alexandre Souza

We’ve received notice that the Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5. We'll begin restoring access tomorrow, and will share an update soon. We’re grateful to our users for their patience, and to everyone who worked with us on redeploying the models.

العملاق الآسيوي Alibaba Cloud يطلق عرضاً هجومياً شرساً ويقدم بنية تحتية سحابية مخصصة بالكامل لتشغيل وتطوير أدوات الذكاء الاصطناعي والـ Agents بتكلفة أقل تصل إلى 50% مقارنة بالمنافسين ​الصدمة هي أنهم يمنحون الشركات المسجلة حالياً حوافز ورصيداً مجانياً يصل إلى 5,000 دولار أمريكي كدعم لتبدأ تشغيل وتدريب موديلاتك بأقل تكلفة ممكنة وبدون مشاكل بطء السيرفرات التقليدية ​إذا كنت تبني مشروعاً يعتمد على الذكاء الاصطناعي وتريد الهروب من فواتير أمازون أو مايكروسوفت الضخمة، فهذا هو الوقت المناسب لاقتناص الدعم وبناء سيرفراتك بنصف السعر ​اطلعوا على رابط التسجيل في التغريدة الأصلية واستغلوا الفرصة ولا تنسوا الإعجاب ومشاركة المنشور لتعم الفائدة على كل رواد الأعمال والمطورين الناشئين في حساباتكم

📣📣 Meet Qwen-AgentWorld — a native language world model that simulates 7 agent environments (MCP, Search, Terminal, SWE, Web, OS, Android) within a single model. Environment modeling is the training objective from day one, not a post-hoc adaptation. 🤔 LLMs are trained to be better agents — better at acting in environments. But nobody has trained them to model the environments themselves. 🗺️ Our roadmap: investigate how language world modeling can push the boundaries of general agent capabilities, along two routes: 1️⃣ Build a foundation model for environment simulation — outperforming Claude Opus 4.8 and GPT-5.4 on AgentWorldBench 2️⃣ Investigate how world modeling enhances agent training: 🔬 Controllable Sim RL (agentic RL with LWM as environments) surpasses training in real environments 🧠 Learning to predict environments (LWM warm-up) makes agents stronger — remarkably, even without any agent-specific training, this predictive knowledge transfers to agentic tasks with zero fine-tuning 📑 Paper: arxiv.org/abs/2606.24597 📖 Blog: qwen.ai/blog?id=qwen-a… 💻 GitHub: github.com/QwenLM/Qwen-Ag… 🤗 HuggingFace: huggingface.co/collections/Qw… 🧩 ModelScope: modelscope.cn/collections/Qw…

DeepSeek-V4-Flash is FREE until June 28! OpenModel Limited-Time Event: → Input: $0.00 / M → Output: $0.00 / M Powerful 284B MoE model with 1M context, excellent coding & agentic capabilities. Other models also enjoy 20%–80% OFF during the event! Try it now before June 28 → openmodel.ai

It’s been few days that @karpathy joined Anthropic team. Funny part is that he can’t use Fable-5 anymore, because he is not a US citizen.

Elon Musk criou, em um único dia, 4.400 novos milionários. Quase 400 deles ultrapassaram os US$ 100 milhões. Não são banqueiros nem investidores de risco. São funcionários da SpaceX: soldadores, técnicos, mecânicos e até funcionários da cantina. Durante vinte anos, a empresa pagou gente de todos os níveis com ações, não só com salário alto. Quem produziu colheu. Juan Hernandez, imigrante mexicano, aceitou um emprego de soldador por US$ 28 a hora em 2015, sem nem saber direito o que era a SpaceX. Recebeu uma pequena participação de US$ 10 mil e pôde comprar mais por desconto em folha. Hoje sua fatia vale US$ 880 mil. Trevor Hise ignorou os conselhos dos pais para pegar um emprego “seguro” na General Electric. Escolheu a SpaceX, ficou 12 anos e acumulou mais de 100 mil ações. Ao preço da listagem, são US$ 13,5 milhões. Aos 37 anos, ele já pode se aposentar. Palavras dele: “A magnitude disso é ridícula.” O detalhe mais eloquente veio antes mesmo da abertura de capital: mais de 100 funcionários se uniram discretamente para contratar uma gestora de fortunas capaz de cuidar de até US$ 5 bilhões. Muitos nunca tinham precisado de wealth manager na vida. Há décadas os IPOs de empresas de tecnologia enriquecem programadores. Desta vez, o dinheiro chegou ao chão de fábrica. Isso é capitalismo de verdade: quem arrisca, quem trabalha e quem entrega valor colhe frutos proporcionais. A esquerda odeia esse tipo de história. Porque ela prova que a verdadeira ascensão social não vem de dividir a miséria alheia, mas de criar riqueza que eleva quem tem coragem de construir.

@ant0ni0_r0mer0 Brand

GPT Image 2 is insane for branding. So I built an agent that generates a full brand kit concept in 60 seconds. Upload your logo, drop a few style references, type something like "make a logo system & variations", then watch the prompt get written and the image generated. Fully automatic. I've been using this a lot recently before moving on any concept, and it's genuinely changed how I approach eCommerce creative production. Brainstorming, concept reviews, pitch decks, it handles all of it. I'm sharing the full agent + workflow. Like + comment "BRAND" and I'll send it over. Need to be connected.

卧槽兄弟们，GitHub 上真的有一堆免费到离谱的项目。 很多能力已经能直接干掉你正在月付的软件。 1. TradingAgents AI 多 Agent 量化交易框架 github.com/TauricResearch… 2. LibreChat 一个界面接入 ChatGPT、Claude、Gemini 等多个模型 github.com/danny-avila/Li… 3. HyperFrames HeyGen 开源的视频生成引擎 github.com/heygen-com/hyp… 4. Fincept Terminal 开源版彭博终端 github.com/Fincept-Corpor… 5. MoneyPrinterTurbo AI 一键生成短视频 github.com/harry0703/Mone… 6. Agentic Inbox Cloudflare 开源 AI 邮件助手 github.com/cloudflare/age… 7. VoxCPM AI 声音克隆工具 github.com/OpenBMB/VoxCPM 8. Flowsint 开源 OSINT 情报分析工具 github.com/reconurge/flow… 9. agent-skills Claude Code 技能库 github.com/addyosmani/age… 10. Nango 开源 API 集成平台 github.com/NangoHQ/nango 兄弟们，这些真不是玩具项目。 很多你还在交月费的软件，GitHub 上已经有人做出开源平替了。 一句话： 别只会收藏 AI 工具网站了。 真正狠的东西，很多都藏在 GitHub 里。

@Zephyr_hg Ideas

50 AI businesses you could start this weekend. Pick one. Every idea comes with a revenue path, MVP scope, and exact tech stack so you're not guessing what to build or how it makes money. Organized into 5 categories: vertical agents, content tools, data infrastructure, edge AI, and services. Medical billing agents. Construction safety cameras. Prompt injection firewalls. Drone inspectors for solar farms. Each one scoped lean enough to validate in 48 hours with a simple ad or DM campaign. No more staring at tech trends trying to figure out where the money actually lives. Pick the lane that fits your skills and ship the tiniest possible version before your competition finds these ideas. Comment "IDEAS" and I'll DM it to you (must be following)

All the best startup accelerators to apply in 2026 (below FULL spreadsheet): [ Top-Tier ] 1. Y Combinator (~$500k, 7% on $125k + uncapped SAFE) 2. a16z Speedrun (~$500k for 10% + $500k follow-on) 3. Techstars (~$220k, bumped from $120k in fall 2025) 4. Founders Inc (~$100-250k for 4-7%) 5. Sequoia Arc (~$1M, terms per company) 6. South Park Commons (~$1M total, $400k for 7% + $600k guaranteed) 7. HF0 (up to $1M uncapped for 5%, repeat founders only) 8. On Deck ODX (DISCONTINUED 2022, skip) 9. Pear VC PearX (~$250k-2M for ~10%) 10. 500 Global Flagship (~$150k for 6%) [ AI / ML Specific ] 11. AI Grant (~$250k for 7%, Nat Friedman + Daniel Gross) 12. AI Fund (~$1M+, Andrew Ng, studio model) 13. NVIDIA Inception (credits + perks, no equity) 14. Microsoft for Startups (up to $150k Azure credits, no equity) 15. Google for Startups AI Accelerator (credits, no equity) [ Vertical / Deep Tech ] 16. SOSV (~$525k across IndieBio, HAX, Orbit) 17. IndieBio (biotech, ~$525k = $250k for 6-8% + Genesis SAFE) 18. HAX (hardware, ~$525k via SOSV) 19. Greentown Labs (cleantech, workspace + grants) 20. Activate (deep science fellowship, 2-yr stipend) [ Crypto / Web3 ] 21. a16z crypto CSX (~$500k for 7%, SF in-person) 22. Alliance (~$500k, ALL18 starts Sept 7) 23. Outlier Ventures Base Camp (~$250k, per-chain verticals) 24. Coinbase Base Builder (varies, mostly non-dilutive) [ International / Regional ] 25. Seedcamp (~€350k-1M first check, rolling, Europe) 26. Entrepreneur First (~$250k = $125k for 8% + uncapped MFN) 27. Antler Disrupt US (~$400k = $250k for ~9% + uncapped) 28. Brinc (~$100k, Asia + Middle East) 29. Station F (Paris campus, hosted-program terms) 30. Founder Institute ($499-999 fee + 2.5% Equity Collective warrant) [ Pre-Seed / Idea Stage / Niche ] 31. Z Fellows (~$10k for 1%, pre-product) 32. ERA NYC (~$100k for 8%, generalist) 33. The Residency (community-first, no standard check) 34. Plug and Play (varies, often non-dilutive) 35. Build For Tomorrow (community + grants) Four things worth knowing: 1. small program acceptance rates are higher than YC, not because they're easier, because the funnel is smaller. apply to 5-7, not 1 2. brand premium on YC is real but not infinite. Arc + Speedrun + HF0 carry signal too 3. "$X for Y%" is the only number that matters. uncapped MFNs are not free money, they dilute you on the next round 4. avoid any list still quoting Techstars at $120k, ODX as active, HF0 at $100k, or Founder Institute as "$10k for 4%". all wrong Shared with you those where I am going to apply with my ideas and products Terms shift annually. Verify on each program's site before applying gl with successful raising

🚨Anthropic recently showed a 24-minute workshop on how to actually do prompts for Claude. Taught by the people who built it. Free. No registration. No paywall. I've seen $300 courses that don't cover what they teach in the first 8 minutes. Watch it and bookmark it now.

A aposta do Sam Altman: uma empresa de 1 bilhão gerida por 1 pessoa, deixou de ser absurdo. Quem ganha em 2026 não usa mais IA. Orquestra melhor. Qual é o primeiro processo que você vai redesenhar?

O humano não desaparece na empresa AI-first. Ele muda de papel. De executor para arbitrador. De operador para auditor. De faz-tudo para pensa-tudo. Aprova, desaprova, critica, analisa. O agente aprende com isso.

Assisti um vídeo ontem (youtube.com/watch?v=pMGg-5…) que mudou minha compreensão de como empresas vão escalar em 2026. IA não é uma ferramenta que você coloca em cima do processo antigo. É um redesenho completo da arquitetura operacional. Deixa eu explicar:

RFS completo: ycombinator.com/rfs Se você está construindo em algum desses espaços, esse documento vale mais do que qualquer pitch deck de tendências.

Para empreendedores brasileiros construindo com IA: Isso é um mapa, não hype. MPEs são exatamente o mercado que o YC está sinalizando. Quem estiver posicionado hoje captura antes de todo mundo.

O @ycombinator publicou o RFS Summer 2026, a lista do que a maior aceleradora do mundo quer financiar. Frase de abertura: "IA parou de ser feature. Virou fundação." O que isso significa na prática 👇