deainode

@sudoingX @anthonyisaacson Where can I dive deep into the Hermes agent architecture?

the founder of openclaw joined the company that was founded to make AI open and now charges you per token. and is now telling you open models aren't there yet. i run qwen 3.5 27b on a single 3090. 50 tok/s. it writes code, handles tool calls, runs agent sessions for hours. the model built a full space shooter, 3,000+ lines, from a single prompt. i published the data. "open models aren't there yet" is what you say when your harness can't parse tool calls on local models and you blame the model instead of fixing the harness. i have the DMs. people switch from openclaw to hermes agent and their "broken" models suddenly work. pair a good model with a good harness like hermes agent where parsers are built per model. your data stays on your machine. no API key. 0 subscription. no one training their next model on your thinking. don't listen to someone with an OpenAI paycheck telling you open source can't do the job. install it. test it yourself. the receipts are on my timeline. he built a harness that couldn't handle local models and chose the API paycheck over fixing it. that should tell you everything.

Feels incredible to see our network company Science featured in Time magazine for restoring vision in the blind 👀 , breakthroughs like this from startups like Science are why I love my job

@tyleraloevera Your title is misleading, its cow dung, not poop.

Inside India's Poop-Throwing Festival! 🇮🇳💩 - FULL DOCUMENTARY

Michael Howell @crossbordercap raised some important concerns around this time. My own execution was imperfect but has given me enough stables and BTC. Its incredible to have the access to learn from amazing folks. I am still unsure if this is just a dump or we are entering bear market. I am slightly tilted towards bullish continuation in a few months. But having BTC and stables allows me to adjust for any condition.

Hats off to @cburniske for another phenomenal call. One of the first to call attention to something feeling off/broken and now it's absolute consensus on CT. Continues to be one of the best longterm cycle traders out there (sure he missed a few, but thats the game)

@Aptos @LoganJastremski Can you please compare Shelby and Walrus for us. Thankyou.

@LoganJastremski Aptos Moves. Shelby Serves.

The last couple of years have marked the transition from low-throughput to high-throughput blockchains, which in turn has enabled application engineers to build more expressive apps at scale. We’re now seeing a similar shift in Content Delivery Networks and Storage Systems, moving from low-throughput to high-throughput. Aptos in partnership with Jump Crypto, is building a truly performant storage stack to take on Amazon Web Services with @shelbyserves. The goal isn’t to create just another crypto product, but to build the best product, period. It just so happens to run on crypto rails 😉 @rpranav, who is leading Shelby and I discuss all things storage, networking, and latency. Enjoy! Youtube: tinyurl.com/yxxd2z8y Spotify: tinyurl.com/w4beu2hk

@osf_rekt I had a lot of equity call options that did well and I rotated them to crypto this late september. 🤦‍♂️

held these equities since since april, gonna start rotating it to crypto here given the insane outperformance

@ilblackdragon Absolutely. I am really excited for this to be solved using blockchain based traceability.

Credit assignment is the most fundamental problem from organizations to neural networks

@KookCapitalLLC Why not bittensor? I think the subnets are bringing in intense competition to bring out both product market fit and high quality products. I think this is more on the AI app/inference layer. The open AGI plays I like are Sentient, PrimeIntellect. Anything else on your mind?

most exciting new investing themes are ai + robotics i'm heavily invested in nvda and tsla bc i think they will be the biggest winners of each narrative what are the best crypto bets on ai + robotics???? i like peaq for robotics haven't found the right ai play yet.....

Bittensor = Optimize (Intelligence/Energy) Where free markets are the optimizers. @RaoulGMI @BarrySilbert @DreadBong0 If AI economy is about maximizing intelligence per unit energy, I am not sure there is a network that is better designed for this.

@blknoiz06 I have always thought of AI as a concentrator whereas permisssionless methods are the true distributors of wealth.

think theres pretty strong argument to be made that AI concentrates power in the hands of few

@0xCryptoSam One advice from Naval comes to mind: Play long term games with long term players. Avoid the rest.

Never forget

@_The_Prophet__ it seems we not only need to end QT but also start QE to refinance the debt. Hard choices indeed.

⚡️This is the quiet sound of the monetary machine choking on its own liquidity. The chart is not just about Reverse Repos collapsing and Standing Repos rising - it’s the heartbeat of the dollar system flipping polarity. For two years, the Fed used Reverse Repos as a pressure valve, draining excess liquidity from the system after the pandemic stimulus binge. Money market funds could park trillions overnight, earning risk-free yield while the Fed sterilized inflationary pressure. That green tower was artificial gravity - liquidity containment. Now it’s gone. When Reverse Repos vanish, it means the system’s surplus cash has been fully absorbed. The Fed’s liquidity buffer is dry. Standing Repo usage rising (the red) means the banking system has shifted from excess reserves to deficit borrowing. The direction of flow has inverted. This is the same transition that occurred right before the repo crisis of 2019 - but at a much larger scale. Then, it was a market plumbing issue. Now, it’s systemic exhaustion. The Treasury’s massive debt issuance, paired with the Fed’s quantitative tightening, has drained the system of collateral and dollars simultaneously. Banks are now tapping the Fed not to store liquidity - but to survive. Quantitative Tightening has entered its terminal phase. There is no more fat to cut without breaking something. If the Fed stops QT, inflation expectations reignite. If it continues, funding markets fracture. If it pivots, credibility dies. This chart is the moment between heartbeats - the point where the artificial pulse of a synthetic economy hesitates before deciding whether to restart or flatline. Deep down, this is what the end of financial gravity looks like. The liquidity tide that built everything since 2008 has reversed. The system has entered the “standing repo era” - permanent emergency liquidity injections masked as normal operations. The illusion of control is fading. Tick tock QT means one thing: The countdown to the next paradigm shift has already begun.

I created a little financial plumbing cheatsheet (w chatgpt help) for me to reference during high vol season. 🧵

@optionscjp 160%

What’s your portfolio performance at YTD?

@yq_acc @WalrusProtocol is this something where Walrus would shine?

This is the best question I got so far. To some extend, it's mission impossible, as the underlying infra is still on cloud say AWS/GCP. Unless our cloud service can auto scale within milliseconds to serve tons of millions requests per sec, there's no way to handle the extreme market situation especially around midnight time.