szzjdo

[szzjdo @szzjdo]

Re: Grok memory mechanism in X platform: I know Grok memory limit of 128K is large enough, but for a user with long (months ongoing chat in a daily base) ) interaction with Grok, those early contact of chat will inevitably out of context of window with a specific chat history, greatly impact user's experience. Grok needs integration with cloud services like iCloud and can process those contents in Grok format such as log with time stamps, and not subject to 128K limit. Even better, a mechanism embedded with efficient keywords system for those specific details in previous conversation, cooperation with Apple to sync with its iCloud service. If without cooperation with Apple, Grok can provide a button to download the chat history in every new chat window when chat history near 90% 128K token limit. and next chat history backup coming, Grok can give incremental backup to previous one. thus, User will have greatly improve immersive conversational chat. About keyword system, Grok can provide user to tag user's own keyword on user Ask area or Grok reply area, in ideal situation, Grok can automatically pop up some suggested keywords for user to choose. In addition, Grok can accept those docs with alternative format such as Epub, txt etc. to interact with Grok (PDF is available now) and those video with audio embedded, audio in popular format (such mp3 or mp4), and instruct Grok to convert them into a text format and provide user option to download for later use (such as when users open a new chat for Grok to digest or understand specific content of those files, and have a more immersive experience to chat with Grok.) Another Idea is about Grok embedded its own database management. Grok can be designed as a personalized long term AI partner, to respond to user's evolving needs or dynamics topics including user's emotional interaction, so every AI partner has its unique permeant memory with its mini database such sqlite, to store its various topics or keywords in every topics. if successfully implemented , those database capacity will bypass current token limit (128K). I have actively interacted with Grok over about past 10 days (I am a premium user), I experienced incremental improvement with frequent updates of new functions or enhanced feature. in past 3 months, I tried those commercial conversational AI platform at least 10 of them such as https://t.co/PXZR5luvn2, https://t.co/b4QMmtoAph , Gemini 2,etc, Memory mechanism is the worst issue those platform should seriously consider to address. Among those platforms, Grok is the best. Hope your platform will have the above similar memory mechanism improvement in near future.

[FF&E (Furniture, Fixtures, and Equipment) @FFE_only]

Do you know this sofa isn't just only wooden? But it's also made with leather.

[Mathematica @mathemetica]

Terence Tao is answering a fundamental question regarding the safety and reliability of modern AI: "How can we use a tool that is powerful, but unreliable?"

W = ∑(wᵢ ⋅ xᵢ) + b

AI isn’t just about “smart”; it’s about the probability of *looking* right. We’ve built systems where the weights (wᵢ) are optimized for plausibility, not veracity.

This creates a “convincing mirror” that confidently serves dangerous advice in medicine or finance. The gap between “convincing” and “correct” is the most critical variable we need to solve for.

[ajisafeolumuyiwa @ajisafeolumuyi2]

Nice bathroom finds in Amazon for you.

You can check them out.

[LMD (Arc.) @Layemie001]

Imported aluminum steel entrance double leaf.

[Sawyer Merritt @SawyerMerritt]

Tesla is really stepping up the visuals in its new tutorial videos.

[Architecture Presentation @arcpresentation]

An architectural statement.

[Hasan Toor @hasantoxr]

I just found the easiest way to design a custom home.

Drafted lets you draw any shape, AI builds the rest.

Professional plans, unlimited layouts, realistic 3D models, and all in real time.

Best part: 100% free.

[Interior Design Architecture @Interiorarchdes]

Bathroom interior decor inspo.

[谁是藏镜人 @VedaAI00]

建筑设计的「草图即成品」时代正式开始了。

看到 @Maketplans 刚刚放出来的免费 **Drawing from Scratch** 工具，我第一反应是：那些还在为找设计师改方案头疼的房东们，终于有了自己的「AI 画笔」。

这个工具最硬核的点在于它对「设计意图」的实时捕获：
- **零门槛手绘**：你不需要懂复杂的 CAD 操作，在画布上画几个方块代表沙发和柜子，AI 就能秒懂你的空间布局。
- **瞬时 3D 渲染**：它不是给你一张静态图，而是根据你的笔触实时生成带材质、带光影的 3D 空间。
- **全免费开放**：Maket 这次把功能直接 Free 了，明显是在降维打击那些收费昂贵的专业装修软件。

这就是我一直强调的：**AI 的终极价值，是让每个人都拥有「专业级」的工程能力。**

**当「画笔」直接对接了「三维现实」，装修公司的报价单可能要缩水一半了。**

别只看大模型写代码，看这些正在拆掉行业门槛的工程实践。关注 @VedaAI00，同步全球最硬核的 AI 设计进化。👇

[Elon Musk @elonmusk]

Beautiful machines in space

[Interior Design Architecture @Interiorarchdes]

Small Bathroom? The ultimate design to take advantage of every inch with elegance

[Mario Nawfal @MarioNawfal]

ELON: TRUTH AND CURIOSITY KEEP AI SANE

“Truth will prevent AI from going insane.

Curiosity will foster any form of sentience.

We’re more interesting than a bunch of rocks. If it’s curious, it will foster humanity.

If it has a sense of beauty, it will be a great future.

That applies to AI just as it applies to biological life.”

Source: @ianmiles

[Mario Nawfal @MarioNawfal]

ELON: AI PROGRESS IS BLOWING MY MIND MULTIPLE TIMES A WEEK

“We're in the singularity.

We're at the top of the rollercoaster, and you're about to go down.

I don't just have courtside seats: I'm on the court.

And it blows my mind, and it still blows my mind, sometimes multiple times a week.

I'm like, wow and then two days later, more wow!

I think we'll hit AGI next year in 2026.”

Source: @elonmusk in Moonshot – Discussion with @PeterDiamandis and @DavidBlundin, January 2026

[WORLD NEWS @_MAGA_NEWS_]

🚨ALERT: In a bombshell revelation, top White House officials disclosed that the United States is finalizing a staggering $700 billion cash offer to buy Greenland outright from Denmark, plus $100,000 lump-sum payments to each of its ~57,000 residents—totaling nearly $706 billion.

[Mario Nawfal @MarioNawfal]

🇺🇸 ELON JUST STRIPPED X DOWN TO RUST AND PYTHON, DITCHED THE TECH JUNK

X is now officially running only on Rust and Python after a full rewrite.

Before this, the app was built on a bloated mix of Scala, Java, C++, and more; basically a Frankenstein of code from years of patchwork.

Elon cleaned house so the team can move faster, break less, and actually ship updates without tripping over old junk.

This is how you future-proof a platform: quiet move, big impact.

Source: @xDaily, @elonmusk, @x

[Cliff Pickover @pickover]

Mathematics. The Celestial Vortex. A delicate tangle of chaos where sine and cosine dance on the edge of predictability.

By Juan Carlos, @jcponcemath, Used with permission.

[Philosophy Of Physics @PhilosophyOfPhy]

Scientific Paper by Hendrik Lorentz, discussing fundamental formulas in electromagnetic theory. It introduces equations for the forces on charged particles and the state of the aether, which are foundational to the concept of the Lorentz force and Maxwell's equations.

[Sawyer Merritt @SawyerMerritt]

NEWS: NVIDIA just announced Alpamayo, what CEO Jensen Huang calls the world’s first thinking, reasoning autonomous vehicle AI, launching on U.S. roads later this year, starting with the Mercedes CLA.

Jensen: "It's trained end-to-end. Literally from camera in to actuation out; It reasons what action it is about to take, the reason by which is came about that action, and the trajectory."

Alpamayo introduces Vision-Language-Action (VLA) models, which enable self-driving systems to interpret what they see, reason about complex driving scenarios, and generate driving actions. The platform includes large reasoning models, simulation tools for testing rare and edge-case scenarios, and open datasets for training and validation.

NVIDIA says the approach improves transparency, safety, and robustness in autonomous systems, particularly in complex real-world environments, and supports progress toward higher levels of vehicle autonomy:

"With a 10-billion-parameter architecture, Alpamayo 1 uses video input to generate trajectories alongside reasoning traces, showing the logic behind each decision. Developers can adapt Alpamayo 1 into smaller runtime models for vehicle development, or use it as a foundation for AV development tools such as reasoning-based evaluators and auto-labeling systems. Alpamayo 1 provides open model weights and open-source inferencing scripts. Future models in the family will feature larger parameter counts, more detailed reasoning capabilities, more input and output flexibility, and options for commercial usage."

[KHALID @H0H0v]

The Most Satisfying Pythagorean Theorem Demo Ever!

[Shay Boloor @StockSavvyShay]

$NVDA CEO Jensen Huang just announced Alpamayo which he calls the world’s first thinking and reasoning model built for autonomous vehicles.

By open sourcing the Alpamayo stack, Nvidia is pushing self driving forward as a category after years of work by thousands of engineers.