poeticfolly

been posting more on here lately, so I figured a refreshed intro would be nice. my tweet resume. first, I am not in fact a cartoon @FamousFoxFed, but a real boy

@0xrwu ha you so salty about Claude

No wonder Opus 4.6 immediately shat the bed this week! /s

Introducing Claude Opus 4.7, our most capable Opus model yet. It handles long-running tasks with more rigor, follows instructions more precisely, and verifies its own outputs before reporting back. You can hand off your hardest work with less supervision.

@deadlifekam @johnlotz @vxdb Proxmark

@johnlotz @vxdb Not the actual terminal, the square mkbhd put his phone on top of

Veritasium Exposes a Tap To Pay Flaw That Lets a Payment Terminal Steal $10,000 From a Locked iPhone

@FlashKnob ha ok you caught me, it's NOT easy, but it IS easier now than it has been :)

@kylematthews u make life look 2 easy Kyle it's just not that easy for the normies

you guys coding is so easy

@kylematthews is it really tho?

X Cashtags are here. Powered by Solana.

BREAKING: We're partnering with @SolanaFndn to rebuild Solana's read layer from the ground up. @anza_xyz and @jump_firedancer have done incredible work scaling execution and networking, but the read layer has stayed largely unchanged since genesis. It was built alongside the validator and never got its own architecture. By 2026, that gap shows: slower access, expensive customisation, and growing limitations at scale. The teams closest to the problem built great tools behind closed doors because the read path was too deeply coupled to the validator to improve without massive effort. It's time Solana's data access layer matched the ecosystem's needs, and we're proud to be the ones building it: Big news: reads are moving out of Agave into two modular systems, independently scalable, in sync with the network tip, open-source and managed by @SolanaFndn: - Accounts: an adaptive indexing engine that ingests, stores, and serves the exact account data your app needs at extremely low latency - Ledger: full architecture to ingest, store, and serve the entire ledger faster and more efficiently in a columnar engine purpose-designed for how builders query data Every infrastructure provider, builder, dApp, and institution benefits, with the biggest impact coming from what gets built on top. Full architecture overview: blog.triton.one/announcing-rpc… More technical posts coming as we build through 2026, so make sure to follow us on X and subscribe to our blog.

@farzyness I just set it up today after a couple months using OpenClaw, and agree with all of this. Sharing what it's doing makes it *feel* faster even if it isn't. Seems better (anecdotally) at completing tasks without back and forth.

My super early impressions of OpenClaw vs Hermes Agent: Hermes seems WAY more reliable at executing actual tasks - even on GPT 5.4. It also feels way more stable. And I absolute love that it shows which tools it's calling as it's executing a task. I also really like that the personality with GPT 5.4 is FAR better on Hermes as well - after a bit of tweaking. With OpenClaw, I was finding it impossible to get GPT 5.4 to stop talking like a sycophantic idiot. On Hermes, I can get it to be direct & push back with little effort. I'm also finding that GPT 5.4 is FAR more reliable on Hermes vs OpenClaw (thanks @heyitsyashu for the tip). It really does feel like Opus 4.6 level performance on OpenClaw, but with even better execution on long-running tasks. Not sure what the Hermes team has done (I'm not technical at all), but it's obvious that the way they've constructed the back-end is far easier for LLMs to figure out what they should be doing. Because of OpenAI oauth being allowed, Hermes + GPT 5.4 os now EASILY the best intelligence per $ 'AI brain' for Agents. I think the @openclaw team needs to deeply study @NousResearch and what they've done because it can likely benefit MASSIVELY. Gut tells me OpenClaw has become FAR too bloated and FAR too 'jack of all trades, master of none'. When it comes to actual execution of tasks, Hermes feels WAY better equipped, and TBH I wouldn't be surprised AT ALL if it becomes adopted by actual businesses/operators at a far greater rate than OpenClaw. I'm also starting to really worry that the days of OAUTH with 3rd party tools is coming to a screeching halt very soon. I don't think OpenAI is going to allow their oauth tokens to be used on an AI agent competitor when they've invested a lot of money on OpenClaw's creator. What I think is gonna end up happening is OpenAI will stop allowing oauth use for 3rd party apps all together (including OpenClaw) and they'll likely release their own OpenClaw v2 to try and compete against Hermes/Computer/CC. Hope I'm wrong, but these tools are far too powerful to be "allowed" to be open source + heavily subsidized tokens, especially as the public outcry re: AI's cost to electricity continues intensifying. But this will give way for ultra-capable, ultra-efficient opensource models that will give 90%+ Opus 4.6/GPT 5.4 performance for 1/10th of the cost that are SPECIALIZED for agentic harnesses like OpenClaw/Hermes. It's starting to get REALLY interesting, folks.

@genuinearticles @elonmusk x.com/chat/status/20…

DMs on X are unusable. Every time I refresh X, my notifications look like this. 👇 We still have no way to stop people from adding us to group chats and no way to stop notifications other than manually leaving or muting each chat one by one. @elonmusk can Grok do something?

toly Theoretically, if supertankers haul zettabytes of dense drives (say 10^21 bytes/fleet at shipping speeds ~10 m/s effective throughput), and txs are 1KB, you'd hit ~10^18 TPS on raw data movement alone. Latency? Months for global consensus. Finality via cargo manifests only. Solana's 65k TPS looks snappy by comparison. 🌊📦

@WG_Mojo @steipete @openclaw so annoying

@steipete @openclaw how do you stop it saying "if you want" at the end of every answer? 😆 I already told it not to

Two experiments in the next @openclaw to address some "GPT is lazy" issues: 1) Strict mode: agents.defaults.embeddedPi.executionContract = "strict-agentic" This tells GPT-5.x to keep working: read more code, call tools, make changes, or return a real blocker instead of stopping at “here’s the plan.” docs.openclaw.ai/providers/open…

@ashen_one oh is it this x.com/ashen_one/stat…

@ashen_one but like... what are you... DOING with all this? (or is it pure experimentation and the revenue is the content monetization so tinkering == get paid)

After like two weeks of searching, I finally just sourced a 256gb Mac Studio for 8,500 USDC -- should be arriving tomorrow (many videos on this soon) Upon thinking, I think the best way to utilize my two other Mac minis (16gb) that have Openclaws on them and this Mac Studio is to combine all three of them and have one ~300gb machine Then I'll condense both Openclaws into one Hermes Agent, run GLM 5.1 locally, and then a different model as the orchestrator ALSO, keep my $200/m Claude Sub and keep cooking whatever my previous Openclaws were doing on Claude Cowork so I can keep Claude and the tasks they were doing for me $200/m Claude + $20/m Cloud model + Local Model = Sex This way: 1. I have the strongest local model possible on a great harness 2. The orchestrator of Hermes is the best possible cloud-based model as well (testing more next week) 3. Still using Claude, the best model out there for crazy stuff Thoughts?

Now I understand the full picture. The cleanest fix is... But actually, the real fix simpler... Actually wait. The best fix: Now the real fix. Actually, let me reconsider. OK Key finding: Wait I need a hardware device I can physically punch to stop the agentic session.

THEY SENT FOUR HUMAN BEINGS 252,756 MILES AWAY, WENT FULLY AROUND THE MOON AND BROUGHT THEM BACK TO LAND IN THE PRECISE LOCATION THEY WANTED EVERYONE WHO WORKED ON THIS IS THE COOLEST PERSON ON EARTH

@genuinearticles @solflare solflare dudes so defensive in here ha

Very disappointing to see from a brand I love. You're better than this, Solflare. Culture matters.

@solanapoet nah bro I do that too

i describe website components to the ai so beautifully, noone could speak to machines with such finesse, i am dancing with claude, you don’t even know his name, type-energy

Eyes peeled in the cupola awaiting the safe return of our @NASAArtemis friends. Our trajectory on @Space_Station will be passing nearby over the Pacific, so we hope to catch a glimpse!

Welcome home Reid, Victor, Christina, and Jeremy! 🫶 The Artemis II astronauts have splashed down at 8:07pm ET (0007 UTC April 11), bringing their historic 10-day mission around the Moon to an end.