Here is a 6 minute call within JOSIE-Voice on a MacBook Pro M4 Pro using: Qwen-3 0.6B TTS/ASR 4bit and Deepseek 4.1 Flash on ollama cloud. it does 100+ Tool calls with agentic work in the real world. No cuts or speedup's.

There is a direct correlation of when I wear my posture corrector and how better my life gets I will not elaborate any further

DHH 这趟中国行，一边被中国汽车震撼，一边嫌中国软件太丑。 在他眼里，中国汽车的豪华程度已经能超越奔驰 S 级，但微信、支付宝、华为鸿蒙这些中国科技的代表，软件设计却依然粗糙，远不如硅谷。 中国硬件都快卷上天了，软件审美怎么还没跟上？ 同一天，他先发帖说，在深圳打了一辆从没听过牌子的中国车，内饰堪比甚至超过顶配奔驰 S 级，20 分钟车程只花了 6 美元。 转头，他又点名微信、支付宝和华为鸿蒙：很尊重中国的 AI 模型和硬件，但为什么软件用起来还是这么粗糙？ DHH 对软件设计有自己非常鲜明的审美标准。毕竟，Omarchy 就是他个人独特的、近乎米其林级别审美的体现。 评论区有人不服：微信把聊天、支付和各种服务整合在一起，办事这么方便，到底哪里不好？ DHH 说，功能齐全不等于设计优秀。他在意的是审美、操作流程和细节打磨：内外边距不一致，页面转场生硬，动画粗糙，信息架构和操作路径也不够讲究。在这些方面，他认为硅谷的软件仍然明显领先。 有人解释，中文能在有限空间承载更多信息，超级 App 又要包办那么多事情，设计精致度自然得让位于功能覆盖面。 DHH 不买账：这是一个虚假的取舍。功能多，不意味着设计就必须粗糙。 希望他这番还算中肯的批评，不至于疏远了他在中国的拥趸。 至于我个人，当然审美有限，还是很喜欢中国的 App。功能丰富，用起来也确实方便。 你同意 DHH 的看法吗？

I'd never heard of the Chinese car brand that just drove us back to the hotel in Shenzhen, but the interior was as nice or nicer than a top-spec Mercedes S class and the 20-minute drive was $6. Germany should be very worried indeed.

Fun challenge video I made about Making Solar Panels with a Microwave :P Seeing what can be done outside of a conventional fab in just a garage with the bare minimum. semiconductor.diy

DeepSeek v4.1 Flash 1017 is basically GPT 5.6 Sol at home Incredible

Here is the next-gen Tesla Roadster development video from the unveil event that they shared exclusively with us.

WHAT THE HELL JUST HAPPENED???😳😳😳

You can pick up a lot of good intuitions about the nature of neural nets from watching this video carefully a few times

🔥 WOW!! The new Tesla Roadster just did 0-60 in 0.81 seconds and Elon says the production model will be even faster!

BOOM! Well, I'll be damned if that's not the very BEST-sounding, 100% simulated/synthesized concert grand piano I've ever heard! FrankenSim.org/music What does that mean? ZERO recordings of real-world sounds were used (other than for comparison and diagnosis), nor were any measurements captured by microphones or similar. Literally everything from start to finish here was part of an enormously intricate physics simulation of a piano: its geometry, materials, moving parts, and acoustic radiation. Oh, and I did the same thing for a full drum kit and cymbals. The cymbals are modelled using the estimated geometry and dimpling pattern of a real Zildjian splash cymbal, together with estimated profile details and literature-based properties for B20 bronze alloy. For the most extreme audio quality, you can pre-render to WAV files or similar. But after many, many rounds of iterative performance optimization, I now have the wasm version running at better than real-time, so it sounds really good live in the browser as well! This is a big deal. It's shockingly hard to get truly realistic-sounding audio from synthesized instruments. The real world is just too complicated. What people usually end up doing is spending hundreds or thousands of dollars to buy these massive libraries of recorded sounds of a real piano, which takes up many gigs of space and is super slow to use. The worst part of that approach isn't the size of the files, the cost of the sample packs, or the inelegance of the approach. It's that EVEN THAT is a massive simplification, because they are recording each key separately at various key velocities and so forth. That means they are missing the very complex interaction effects that arise in a real piano when you play certain keys at the same time or after one another. The energy persists in the instrument and conditions the next sounds. It's subtle but real and you pick up on it when it's missing. And yes, they do try to capture some of this, but they use very lossy methods with more simplified approximations that are way more primitive than what I'm doing here with FrankenSim. I've invested a massive amount of time and focus on ensuring that the energy accounting is correct in these simulations. And the same thing goes for drums! There are complex interaction effects. These sorts of things are emergent and "just fall out" when everything is simulated for REAL using the actual underlying physics. So why haven't other companies and people done this? I'm not actually sure; it's kind of insane when you consider all those people working at Yamaha, Roland, Korg, Moog, etc. But my guess is that it's because it's super hard and complicated to get all the physics right and to program it the right way so that it's tractable without needing crazy amounts of compute. Anyway, take a look at this demo site I set up. It's a lot of fun to play with, and at the bottom there is a nifty MIDI player that takes any MIDI file and lets you listen to it with the high-quality simulated piano and drums in the browser. This is the culmination of months of work and hundreds of billions of frontier tokens. It all started when I wanted to see what Fable 5 could really do, so I thought of the most comically ambitious project I could think of: FrankenSim, the system that unifies ALL of physics into a single coherent and cohesive simulation across all domains: mechanics, fluid dynamics, electricity and magnetism, etc., instead of modeling each domain as a separate silo, where each domain can use the same generic geometric primitives, solvers, calculus machinery, etc. All unified using the language of sheaf cohomology to maintain consistency between representations, and where everything is made as legible, intuitive, and ergonomic to AI agents as possible. This music simulation stuff is just the beginning. Beyond many more instruments, I'm going to be using FrankenSim to model HVAC systems, fluid flow through plumbing systems, and more!

really feels like AI may be on the cusp of something now that's it's reverse engineering things.

after switching our approach to building a "FirmOS", I'm increasingly confident that the future of agent collaboration within organizational settings is going to be a "company operating system" or CompanyOS, designed for agent and human collaboration instead of a "company brain" that holds everything, the OS is for storing rules, processes, decisions made, which tools to use and how. it can mirror some data, but the source of truth is often best held in your existing tools (CRM, etc.). it's continuously updated and used by your agents i think of this as a "flipped playbook" where instead of writing down the company rules first and following them (which is kind of like waterfall development), you're constantly updating the playbook as you decide or learn new things day-to-day running the business

whoever designed humans was definitely a vibe coder

iPhone 18 Pro series logic board is comically small, even with WMCM. Making a board this dense takes pretty much every cutting edge tech in the industry.

This would be Claude’s mental model for life

Argon has this amazing natural understanding of the world. It can do so many things from first principles. Here's a pelican riding a bike, using fourier epicycles. Video made by Argon. Time-synced Lyria 3.5 music prompted by Argon.

I built Ctrl+F for X-rays 🔥 type "bone" and SAM 3.1 from @AIatMeta finds 50 of them on a hand X-ray in 286 ms. then "finger", "hand", and on a surgical tray, "scissors" and "hammer" ALL LOCAL! only one NVIDIA A100 on @huggingface 🤗

A day after Luckin Coffee admitted inflating its sales by around $300 million, Chinese customers waited in an hour-long line to buy its coffee. It was April 3rd, 2020, and by noon, the company's app crashed. Lu Zhengyao, the chairman, posted on WeChat, "Today, be even more full of energy." One part-time barista said that, on an average day, her store fulfilled some twenty orders. That day, she and her manager fulfilled more than a hundred and seventy orders, nearly a third of which were free, thanks to Luckin's promotion for first-time customers. With so much going out free, it made her feel like she was "paying off a debt," she said. Another barista, in Beijing, said that some customers came to the store to support Luckin, "the pride of domestic brands": the investors who lost money were American. The scandal began two months earlier, when the short-sellers at Muddy Waters published an anonymous report based on physically counting customers. The firm hired over a thousand and four hundred part-time workers to visit six hundred and twenty stores and press a button on a clicker every time someone entered the store; some pretended to be "students writing a paper." Every Luckin order gets an incrementing pick-up number, so the short-sellers placed orders when the stores opened and when they closed. The gap between the two numbers was seventy-two percent higher than the number of customers they saw. (In November, store managers received a notice that the pick-up numbers were going to "increase randomly.") When China's market regulator finished its investigation, it found one hundred and twenty-three million fraudulent orders for coffee vouchers. Luckin had written checks for supplies and advertising that it never bought, then booked the money when it was repaid as sales. Luckin's stock, which was listed in New York, plummeted by three-quarters in a day, even as customers continued to line up for subsidized coffee. But its customers never left. Under the private-equity fund Centurium, Luckin launched a Moutai latte, in 2023, using China's most celebrated liquor. Although the amount of alcohol in each drink was small, around two milliliters, the dairy supplier said that Moutai sent people to watch as the bottles were poured, one by one. Luckin sold five million, four hundred and twenty thousand cups on the first day, and by the end of the year it overtook Starbucks' revenue in China. The company now has over thirty-six thousand stores, and last year it generated around seven billion dollars in revenue. In May of this year, Centurium, which controls Luckin, acquired Blue Bottle, the popular American coffee company, from Nestlé. Moutai bottle image: Jula2812 / Wikimedia Commons / CC BY-SA 4.0.