sungang

null

[Phosphen @phosphenq]

https://t.co/euPsjfIvce

[GSR @GSR_io]

This is a must read.

[Cred @CryptoCred]

Good morning, fellow Web3 digital asset investors

Which highly correlated cryptocurrencies with unsustainable emissions, no structural bid, and a black hole of passive drift to the downside are we buying today?

[谁是藏镜人 @VedaAI00]

把干巴巴的 CS 算法变成了 300 多个可交互的「视觉沙盒」。

Elijah Yilma 发布的《The Interactive Handbook on Data Structures and Algorithms》不仅仅是一本电子书，而是一个可以直接在浏览器里运行代码的「学习操作系统」。

这背后的工程设计非常惊艳：
1/ 「空间可视化」。数组、哈希表、树结构不再是抽象概念，你的每一行操作都能在内存视图中看到实时的几何反馈。
2/ 「原位执行」。内嵌了完整的 Python 运行时，边看边写，无需切换窗口。
3/ 「视觉调试器」。支持算法执行轨迹的时间回溯，你可以像看电影一样随时暂停、快进和倒放代码逻辑。

计算机科学的学习门槛，正在被这种极度硬核的互动式 UI 彻底打碎。

我是「谁是藏镜人」，专注于挖掘 AI 与交互设计演进的底层逻辑。
关注 @VedaAI00，带你看清教育与技术交汇的每一个进化信号。

[Killa @KillaXBT]

$BTC

The funniest part is how this plays out the same every single time, yet people still say "this time is different."

Friendly reminder for those who don’t understand how the news algo works:

Good news in an uptrend > either continuation or a local top, depending on price action, extension, re-accumulation, etc.

Good news in a downtrend > temporary boost to give market participants optimism before price starts dumping again.

Bad news in a bear market > usually leads to extended downside.

Bad news in a bull market > mostly just FUD, and price pushes higher anyway.

It’s all an algo. Study it. Learn from it. This is how they trap the herd.

[郭宇 guoyu.eth @turingou]

最近我已经用 AGS（Agent Git Service） 替代了 GitHub 作为我的主要开发环境，年初我曾经想过彻底放弃本地开发环境，5月底终于做到了，现在，除了 iOS 模拟器依赖 Mac 我不得不让我家里的 Mac Studio 始终在线之外，其他产品的开发我已全部迁移到围绕 AGS 的第一层 harness 工作流，因此，AGS 成了我的 source of truth，GitHub 仅用来做代码备份和开源项目的 issue 管理。

我的 AGS 开发服务器运行在 sandbank cloud 的第一台服务器，也就是东京（TYO）服务器上，工作流是截图中第二条。dev agent 会自动在 AGS 服务器上进行开发，支持 wanman 和 codex /goal 双驱动的模式，提交，构建，stage 环境和发布都在本机。至于 chat UI，我现在选择 chatGPT ios app 上的 codex 来监控和指导对话，没有用其他聊天软件。

基于 AGS 的工作方式，我在做一个新的产品叫做 VAS（virtual agent server）除了整合上述功能并提供虚拟化的工作空间之外，VAS 还提供了双工语音模型的控制入口，这样我们就可以把这样一个「可以在生产环境开发真正产品」的 harness 整合到任何音箱或者语音人机界面中去。

[melon_melon_head @MelonConnsoieur]

@SalsaTekila Yeah don't run my code if you'd like your vibe coded setup to use it to build it's systems I would suggest just letting it have a look . Even with cc between testing and ironing out kinks for the strats to work it took me a month

I completely get it!

https://t.co/wQ1w1IMmHH

[Capital Flows @Globalflows]

Every economy in the world has a very clear choice right now:

1) hike and push their economy into a recession, in which case China hollows out more of that country's industrial base

OR

2) Blow up this bubble even further

Every country will choose #2. You would do the same if you were in their shoes

[Capital Flows @Globalflows]

The current credit cycle melt-up is being driven by two things:

1) Financial market liquidity is expanding AT THE SAME TIME credit is being injected into the underlying economy. This creates a reflexive feedback loop between the market and economy as they feed on each other and risk assets are the release valve of this liquidity and credit.

2) AI is fundamentally retooling the market and economy in a manner that people don't really understand yet. AI is speeding up the pace at which goods and services are transacted in the economy. It now costs a lot less to spin up a company, market it, and build a brand. This can happen with very little upfront capital which means that more businesses generate cash flow with less up front investment. On net this basically injects more cash into the system without being very capital intensive. The same thing is happening in financial markets with capital allocation decisions.

The problem is that there is an increasing amount of money chasing the same amount of investments even though things in the economy and market are operating faster due to AI.

I have been explaining all of these factors daily on the livestreams I do and in the daily reports i send out (both of which can be found on my website which is linked in my bio)

[jeff.hl @chameleon_jeff]

Excited to see everyone come together for this historic moment. AQAv2 brings the protocol-aligned stablecoin model that @Nativemarkets trail-blazed to USDC with @Coinbase and @Circle's commitment to Hyperliquid. The community no longer has to choose between liquidity and alignment.

Our industry will face adversity as we continue to grow. It gives me hope seeing titans of the industry come together to build for users and bring all of finance onchain.

[Capital Flows @Globalflows]

Here are all the main ideas I covered in the livestream explaining interest rates 🧵

1. The rate complex is the operating system of every other market. There is no single interest rate. Fed funds, SOFR, the 2 year, 5 year, 10 year, and 30 year each price a different question and transmit to different parts of the equity complex, the economy, and FX. SOFR and Fed funds price what the Fed will do. The 2 year prices the policy path with marginal duration. The 5 year is the belly hinge between policy and term premium. The 10 year and 30 year are pure duration. Read them as one system or you miss the actual signal.

2. Coinbase capitulating to Hyperliquid via the AQA V2 is the structural tell that PURR added to the US is coming this month. Coinbase ignored Hyperliquid for years, and the AQA V2 announcement with Circle’s commitment to deploying USDC on Hyperliquid is a functional capitulation. PURR is up roughly 20% on the day and Hyperliquid is finally diverging higher from Bitcoin. The setup is not priced in yet..

I have been laying out the PURR thesis for months now here:https://t.co/cI67kjtFih

3. SOFR contracts are pricing 12bps of hikes between now and December 2027. That is the entire forward curve. The Z6 and Z7 contracts have moved from pricing cuts to pricing marginal hikes as inflation swaps tick higher. The Fed is not actually going to hike. They are going to pause into this, which forces real rates lower. The mispricing in the forward curve is the trade. Watching SOFR alone tells you what the entire rate complex is doing because every other rate is downstream.

4. The 30 year nominal is at 5 percent, the 30 year real is at 2.5 percent. Both are at the upper end of their multi year ranges. Despite that, REITs are barely off all time highs, homebuilders are holding levels, and the broad equity complex is at new highs. This is the cleanest tell that the system is not breaking under restrictive rates. The 30 year is doing what it does, and the economy is absorbing it.

5. Short end inflation swaps are higher than long end inflation swaps. The 1 year is at 3.3, the 30 year is below it. That curve shape tells you the market is pricing the current inflation impulse as a near term phenomenon and not a structural repricing. If crude continues to bid and core CPI starts transmitting, the 2s10s inflation swap curve will steepen further. Watch that curve. When it turns inverted to negative, that is the early warning for a bigger inflation cycle.

6. The fact that interest rates are rising and equities are not falling is the cleanest tell of resilience. The system is taking direct rate punches and not moving. XLI, XLF, and the small cap IWC are all holding new high territory while the long end is at the upper range. Industrials, financials, and small caps are the three rate-sensitive complexes, and all three are bid. That tells you the underlying economy can withstand much higher rates than the current setup implies.

7. Real rates are driving the short end. The Fed is choosing inaction into the inflation impulse, which is mechanically the liquidity mechanism. The attribution analysis on the 2 year shows inflation expectations driving the recent move up, not real rates. The Fed is not taking an active stance. They are letting it run. That inaction equals real rates falling, which is the engine of the credit cycle melt up. Once real rates print negative on the 1 year, the next leg activates.

8. The IWC small cap ETF rallying while rates rise is the signal you do not want to short equities. Small caps have the highest sensitivity to the underlying economy and to financing costs. When they rally through a rate impulse, growth is structurally stronger than the rate move can offset. Combined with industrials and financials holding new highs, this is the cleanest cross sector confirmation that the credit cycle is intact.

[hanlon’s mortola razr @rhizomaticthot]

the fast16 malware was almost certainly targeting spherical implosion simulations.

left: unmodified LS-DYNA 970
right: LS-DYNA 970 modified with the relevant portions of fast16.sys

both running a spherical implosion deck

[Kyle @zeroxkyle]

This is a must read for anyone looking to be an operator

There's another tangential principle that I like - at high speeds, going 80km/h vs 100km/h makes little difference to total time taken

It's how long you can go without stopping

https://t.co/XnH7ZNQXU4

[Warren Pies @WarrenPies]

Are we in a bubble? Is this 2000?

IMO, no and no.

Remember, these questions have been in the air for years now (excerpt from a March 2024 report)

Fiscal policy remains one of the most important macro differences
-In 2000 govt was running a 2% surplus
-Today the deficit is ~6%

[CrediBULL Crypto @CredibleCrypto]

When you start hearing one of the largest holders of Bitcoin making statements about how Bitcoin just needs to appreciate by “x% annually” for their business plan to be viable you just know that the first secular bear market in Bitcoin’s history is gonna be a painful one.

The idea that we will see a bear market worse than anything we have yet seen seems like a foreign concept for some, despite the fact that we have seen these multi-decade boom and bust cycles play out again and again in every other financial market.

If you plan based on data from Bitcoin’s past when its entire past has been one massive secular bull cycle, then you’re gonna be completely blindsided when Bitcoin enters its first secular bear- conditions that we have never yet actually seen before.

$BTC

[Macro_Lin ｜ 市场观察员 @LinQingV]

很赞同郭宇这个判断。技术主义者的悲观和乐观从来就是一体的。

过去半年多，我在AI的帮助下完成了一颗以LLM为target workload的推理芯片的端到端设计。参考某家的架构，从high level simulator到C-model到RTL，现在马上进入验证。我之前写过一个半吊子编译器，但SystemVerilog的RTL一行都没写过。

由于没有流片预算，这颗芯片可能永远只能跑在VCS仿真器里。但这个过程证明了一件事，以前需要几个团队协作一年才能完成的工作量，现在一个人借助AI就能覆盖。

反过来想，当AI把设计端的门槛压到接近于零，止步的地方恰好就是价值最高的地方。流片。一颗先进制程芯片的tapeout成本是几千万美元，这个数字AI一分钱都帮你降不了。未来真正的价值锚点就在这些重资产的制造端，先进制程的fab、先进封装、HBM。能设计芯片的人会越来越多，能把芯片造出来的产能永远是稀缺的。

[Capital Flows @Globalflows]

No one looks at these trash consumer sentiment indicies btw

Large traders and institutional money don't care AT ALL

When people say, markets are at highs but consumer sentiment is at lows so that means were going to collapse is looking at a datapoint that is irrelevant

Start with price action and figure out how to predict that instead of starting with broken data

[Simon Dixon @SimonDixonTwitt]

Makes you wonder why you pay taxes

[Antonio Lupetti @antoniolupetti]

AI and Consciousness.

There’s a lot of debate around AI and whether consciousness could emerge from systems like LLMs. It’s a natural question, given how well these models simulate language and reasoning.

This Google paper challenges the idea that consciousness could arise from computation alone. The key point is that computation is a description, a map we assign to physical states, not something that exists intrinsically in matter, and a map (no matter how precise) is never the territory in any real sense.

So increasing complexity isn’t enough to generate consciousness. We may get more and more convincing simulations, but that doesn’t imply the emergence of actual conscious experience.

https://t.co/SXPH14N3Vt

[🐧 @Pentosh1]

I think a lot of what we are seeing today still stems from covid and the animal spirits that created of an entire world be locked inside with stimmies and nothing else to do and we are going to see insane boom and bust cycles inside of AI.

The wealth that has been created from the AI boom in the last year for those who have bet on it has just gone parabolic. And I think in general that will continue for years to come but with added volatility in short boom and bust cycles. But in the end, the believers will win

Seeing entire index's like these be up 2-3-4x in a year is insane bc inside that bc you know individual stocks are up 100x+ in many cases. Then you have the rise of prediction markets, sports gambling, online casinos. There is something for everyone

have said repeatedly I felt the future here is bright. and I think for risk takers looking at new and emerging tech will continue to win. Hard to imagine all of the exciting things that are going to come out of this that we just didn't have the capability to build out before. In so many sectors. Defense, energy, robotics, AI, and probably anything else you can think of.