Jason

🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length. 🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models. 🔹 DeepSeek-V4-Flash: 284B total / 13B active params. Your fast, efficient, and economical choice. Try it now at chat.deepseek.com via Expert Mode / Instant Mode. API is updated & available today! 📄 Tech Report: huggingface.co/deepseek-ai/De… 🤗 Open Weights: huggingface.co/collections/de… 1/n

DeepSeek-V4-Flash 🔹 Reasoning capabilities closely approach V4-Pro. 🔹 Performs on par with V4-Pro on simple Agent tasks. 🔹 Smaller parameter size, faster response times, and highly cost-effective API pricing. 3/n

DeepSeek-V4-Pro 🔹 Enhanced Agentic Capabilities: Open-source SOTA in Agentic Coding benchmarks. 🔹 Rich World Knowledge: Leads all current open models, trailing only Gemini-3.1-Pro. 🔹 World-Class Reasoning: Beats all current open models in Math/STEM/Coding, rivaling top closed-source models. 2/n

The DESIGN.md upgrade is officially live in Stitch! 🚀 Yesterday we open-sourced the spec, and today we are giving you more tools to use it: 📄 Start with DESIGN.md: Attach a DESIGN.md right at the start of a project. Don't have one? Our new wizard will help you extract one from your code base or the web. 🔄 DESIGN.md to Design System: Upload a DESIGN.md file to generate a Design System on the canvas. 🧩 Token Support: Design Systems now support design tokens from DESIGN.md! Add them to your file and Stitch will seamlessly incorporate them. 🗑️ Clean Up: (You asked, we listened) You can now easily delete a design system. We’re excited to build the open design standard together. 👇

🚨CHINA’S MEDICAL LLMs ARE NOW LIVE IN HOSPITALS. there’s 42 LLM powered doctors and nurses across 21 specialties in a hospital in tsinghua. they ran around 10k+ simulated patients through it in just days and hit 93.06% accuracy on MedQA. this usually would take doctors years to process. and this isn’t just a research paper btw. Hainan Boao opened China’s first fully AI native hospital recently along with DeepSeek medical LLMs already running in 260+ real hospitals across the china. - while everyone else publishes benchmarks, China is treating actual patients with it. insane. China seems to be aggressively pushing medical AI in real hospitals faster than most countries.

🚨 Claude just got access to live stock prices, balance sheets, cash flow statements, and breaking company news in real time. No hallucinations. No guessing. Actual structured financial data. It's called Financial Datasets MCP Server and Wall Street should be nervous: ↓

🚨 Breaking: Alibaba just killed the browser automation stack. **page-agent.js** — a GUI agent that lives directly inside your webpage. No Selenium. No Puppeteer. No Chrome extension. No Python backend. Just one script tag. It reads your DOM as text (no screenshots, no multimodal BS), brings your own LLM, and executes natural language commands like "fill out this form" or "click login" — right inside the page. The use cases are genuinely insane: → Ship an AI copilot in your SaaS in literally lines of code → Turn 20-click ERP/CRM workflows into one sentence → Make any legacy web app accessible via voice or natural language 12k stars. MIT licensed. Built on top of browser-use internals — but without any of the setup overhead. This is what "AI-native UX" actually looks like in practice Link in comments👇

Say goodbye to video editors. This open-source tool turns a news headline into a published YouTube Short in one command. It's called YouTube Shorts Pipeline and it chains Claude, Gemini Imagen, ElevenLabs, and Whisper together into one pipeline. The cost breakdown is brutal: Claude script: $0.02 Gemini visuals: $0.03 ElevenLabs voice: $0.05 Total: $0.10 per video Type a topic. Get a live YouTube link. 3–5 minutes. Supports multiple languages. Custom voice IDs. Dry-run mode to preview before producing. Manual script override before rendering. Everything stored locally. No cloud dependency. 100% Open Source. MIT License.

盘口只是表面，数据源才是金矿，如何省下一年数十万美金的终端订阅费 华尔街把这些能力包装成高价终端机，卖你几万美金一年。 但另一边，GitHub 上早就躺着一堆开源仓库，直接把这层资讯壁垒撕开了。 这次我直接帮你整理 15 个顶级仓库。 从美股数据、新闻监控，到总经研判、量化套利，一次打包。 1、华尔街底层数据套利：yfinance 这个仓库让你直接绕过券商的限制，提取 Yahoo Finance 的底层数据。无论是个股历史价格、即时报价还是机构财报，一键抓取。散户还在看延迟盘，你已经自己建好数据库了。 ：github.com/ranaroussi/yfi… 2、全自动投研报告套利：automated-market-report 每天花几小时盯盘写分析？这个脚本自动整合 Yahoo 和美国联准会 FRED 的数据，每天自动生成精美的 PDF 市场日报，直接拿去做付费社群的早报都行。 ：github.com/hgnx/automated… 3、无代码批量爬虫套利：stock-market-scraper 不懂写代码也没关係，这个终端工具让你敲几行字就能把全市场的历史数据扒下来，自动存成 CSV 或 JSON，直接拿去餵给你的 AI 模型做预测。 ：github.com/gunjannandy/st… 4、顶级媒体聚合套利：FinNews WSJ、CNBC、路透社的资讯极度分散且有付费牆。这工具能一键聚合各大顶流财经的 RSS。当宏观经济异动，你永远比大众早半小时拿到一手消息。 ：github.com/scaratozzolo/F… 5、彭博终端机平替套利：realtime-newsapi 金融机构花几万美金订阅彭博，这个开源项目底层对接了上万个新闻来源，直接帮你打造一个免费的实时金融快讯看板，监控全球异动。 ：github.com/janlukasschroe… 6、财经历史大数据套利：financial-news-api-python 涵盖 Barron's、Bloomberg 等 5000 万篇历史财经文章的接口。想回测「某个利空消息出来后股价怎么走」，这个资料库就是你的时光机。 ：github.com/FinancialNewsA… 7、市场情绪量化套利：WSJ_WebScraping_NLP 散户被新闻牵着鼻子走，聪明钱在量化情绪。直接抓取 WSJ 海量文章做 NLP 情感分析，看透媒体在煽动什么情绪，反向寻找交易胜率。 ：github.com/philippe-heitz… 8、付费牆情报挖掘套利：wsj.com-scraper 专门针对《华尔街日报》设计的爬虫工具。那些藏在付费牆后的深度市场洞见和机构观点，透过技术手段直接提取，打破阶级信息壁垒。 ：github.com/pim97/wsj.com-… 9、特定公司狙击套利：Reuters-News-Scraper 想要埋伏特定公司的财报或併购案？这个针对《路透社》的轻量爬虫，输入公司代码就能扒出所有历史关联新闻，深挖隐藏的利好线索。 ：github.com/maximenc/Reute… 10、即时盘口新闻套利：ycnbc 专为 CNBC 设计的即时监控模组。美股开盘时，秒级监控首页头条与联动的市场指数。适合做日内短线，跟着突发新闻做波段的高手。 ：github.com/codestorm-offi… 11、总经数据降维打击套利：graphic-detail-data 《经济学人》官方开源的图表原始数据。别看二手财经 KOL 瞎吹，直接拿顶级智库的底层 CSV 数据做总体经济趋势研判，降维打击。 ：github.com/TheEconomist/g… 12、宏观汇率定价套利：big-mac-data 同样是《经济学人》官方维护的「大麦克指数」。精准定位全球法定货币的购买力平价，做跨国资产配置和外汇对冲，这是必备底牌。 ：github.com/TheEconomist/b… 13、顶级智库思维克隆套利：web-scraping-economists 专为《经济学人》设计的深度爬虫教学。把顶级经济学家的文章全部爬下来，餵给你的大模型，直接克隆一个专属于你的首席经济学家。 ：github.com/mjavadshamsi/w… 14、量化策略白嫖套利：best-of-algorithmic-trading 不要闭门造车写策略！这里有全球最顶级的开源量化框架、回测工具和交易机器人排行榜。把别人验证过的印钞机 Fork 下来，稍微改装就能跑。 ：github.com/merovinh/best-… 15、华尔街宽客武器库套利：awesome-quant 量化金融的终极百科全书。从衍生品定价到机器学习预测模型，里面藏了几百个专业级别的代码库。华尔街 Quant（宽客）在用的技术，这里全公开。 ：github.com/wilsonfreitas/…

Vibe coding is more addictive than any video game ever made (if you know what you want to build).

Harvard made its Senior Engineer roadmap available to the public at no cost. Stop paying for $2,000 bootcamps. Prof. Vijay Janapa Reddi just put the entire ML Systems (CS249r) curriculum on GitHub. If you master these 6 pillars, you're ahead of 99% of the field: 🏛️ Architecture 🚿 Data Pipelines 🚢 Production 🛠️ MLOps 🔋 Edge AI 🔒 Privacy This is the "Black Box" of Big Tech infrastructure, open-sourced. Read. Learn. Bookmark.

ml-visualized.com/index.html • watch algorithms train step-by-step • see gradients, weights, decision boundaries evolve • math + visuals + code in one place

🚨 BREAKING: Someone just open-sourced a full satellite ground station you can run from home. No internet. No expensive setup. Just an SDR… and space above you. This system lets you: 👉 Track satellites in real time 👉 Capture radio signals directly from space 👉 Decode weather images (NOAA, METEOR) 👉 Save raw satellite data to your PC All locally. Most people think this requires: • Huge antennas • Paid APIs • Government-level tech Not anymore. What makes it crazy: → Works completely offline → Uses a cheap SDR dongle → Turns signals into real Earth images → Builds your own private satellite pipeline You’re literally going: Space → Signal → Your Computer Why this matters: We’re moving from: Consuming space data → Owning the pipeline 100% open source. Anyone can build on it.

🥳 AI Update! I was responsible for rigging and animation, in Maya and built a AI pipeline to look at the playblast and add fur, texture, light and render together while staying true to the original motion. youtu.be/WusKkCVu15I?si… #animation #3d #anim #ai #animation

Lmaooo this app idea is genius 😭 They got 2.9M views by literally just screen recording their app This is the strongest form of marketing The app is a gamified pushups tracker Basically an RPG game but you can only deal damage by doing pushups These founders are probably 18-19 Stop wasting your time and get to building that app In the article below I teach you exactly how to do it ( step by step ):

🚨 BREAKING: Chinese researchers just shrunk a robot's brain by 11X. They just dropped "BitVLA”, a 1-bit vision-language-action model that shrinks a robot's AI brain by 11x, so it runs entirely on dirt-cheap hardware. Normally, if you want a robot arm to see an object and know exactly how to grab it, you need a massive model running on expensive GPUs or a constant server connection. This team fixed that by compressing the entire model down to just 1.4GB. They simplified almost all the complex internal math into just three basic numbers: -1, 0, and 1. The crazy part? They proved this extremely compact model still moves the robot just as accurately as massive, heavy models, but it operates four times faster. This means we can now build smart, highly capable robots using low-power chips locally, completely offline. No expensive server connections. No latency.