Heath Barber

But don’t worry.. their no limit accounts let them runs loops that run hours and overnight as he has discussed several times about “loop engineering”. The issue is the lack of practicality, esp from Boris, with the unrealistic spend to execute some of the things he is suggesting. Ofc if your account is in god mode this isn’t an issue.

Fable 5 burned 28% of my weekly limit AND used up my 5 hour limit with two prompts in about 30 minutes. @bcherny @claudeai how is this acceptable with a 20X plan? It’s unusable

@ChipJohnson01 Me

Who in the world has 4 hours and 40 minutes to sit down and watch a YouTube video?

@ForePlayPod Oh darn…

Michael Block with a triple bogey on 18 and drops back to T8. PAIN.

@TheeDarkCircle That would get rid of me too!

Gasoline gets rid of wasps in a matter of seconds

@deftersanat Wish I had that much free time!

Bana şu kadar yer yeterli…

@Jules31415 I thought he was father of the bride.. not the groom!! 😬

Fox News senior White House correspondent Jacqui Heinrich married GOP Rep. Brian Fitzpatrick. 👀 They wed in NYC, which held a special significance as Heinrich's first assignment as a network news reporter and Fitzpatrick's first posting as an FBI agent. Kevin McCarthy and Joe Manchin delivered toasts during the reception.

@Rainmaker1973 My charcoal doesn’t work like that!

The way this charcoal ignites [📹 asidi77999]

@goodgood_golf @GolfChannel Wait they’re playing in a tournament? Hade NO idea. Not sure what that says.

Steve might be the hottest putter in golf right now.. Make that 7 STRAIGHT birdies for him and Alexis 😳

Ideally, yes, this would be incredibly powerful and would intelligently establish and evaluate results or conditions that are broadly established. However, the practicality of this is not feasible at this time. Unless: 1. You have the hardware requirements locally to manage your local LLM(s). 2. You have no fear of RIDICULOUS token cost for auto-renew accounts on your favorite platform. The second option is potentially viable and financially acceptable if the loop solver is simple, but that kind of defeats the purpose of this idea.

Claude Code creator, Boris Cherny: "Every night I have hundreds, sometimes thousands of agents running in loops for 5, 10, 20 hours straight. This is just how engineering is done now." That's the shift that will define the rest of 2026. 53 minutes of pure insight from the creator of the tool every AI engineer now uses. Watch it, then read the full guide on loops below.

@msturgis78 @Hertz This has happened more and more lately.

Hey @Hertz, What exactly is the point of a reservation if you have absolutely no car available when I show up? Your employee couldn’t have cared less. No apology. No attempt to help. No solution. She literally shut off the lights and walked away while I was stranded trying to figure out how I was getting home. You took my reservation, took my time, and then left me standing there like it was my problem. This wasn’t bad customer service. It was complete indifference. Absolutely unacceptable.

SAM ALTMAN WATCHING CHATGPT HALLUCINATE LIVE ON STAGE IS THE FUNNIEST THING I'VE SEEN ALL WEEK. THE CEO OF OPENAI, ON STAGE, IN FRONT OF EVERYONE, WATCHING HIS OWN AI JUST MAKE THINGS UP IN REAL TIME. AND HIS FACE SAYS IT ALL. THIS IS THE GUY TELLING US AGI IS COMING SOON BTW.

@timburchett I’m good

I hate askin’ for money more than a root canal, but here it is straight from me: I need your help.

@BasedMikeLee Ok and he will live the rest of the days stupid rich and smiling. Do something!

Anthony Fauci inflicted immeasurable harm on the American people—all while misleading Congress about gain-of-function research funding, shaping the narrative against the lab-leak hypothesis, and directing taxpayer dollars to risky coronavirus experiments History will revile him

Web scraping will never be the same. (100% open-source visual search at scale) PixelRAG is a retrieval system that skips HTML parsing completely. Instead of scraping a page into text and embedding chunks, it screenshots the page and retrieves the image. A vision-language model reads the answer straight off the pixels. Why that matters: parsing is where web RAG quietly loses information. - A single HTML-to-text parser can drop 40%+ of a page. - Tables, charts, and layout get flattened or thrown out. - Swapping parsers alone can move accuracy ~10 points on the same docs. PixelRAG indexes the page a person actually sees. The team built a visual index of all of Wikipedia, 30M+ screenshots, and it still beats the strongest text RAG baseline by 18.1% on text-only QA. The repo also ships a Claude Code plugin that gives Claude eyes. It lets Claude screenshot any URL and read the rendered page instead of scraping the DOM. So you can hand it a live page, an arXiv paper, or your local site and ask what it actually looks like. One setup script. No MCP server, no backend. How the pipeline works: - Renders each document (web, PDF, image) to image tiles. - Embeds them with Qwen3-VL-Embedding, LoRA fine-tuned on screenshots. - Builds a FAISS index and serves a search API. A stronger reader model lifts accuracy with no re-indexing, since the index is just pixels. Everything is open-source under Apache-2.0. GitHub repo: github.com/StarTrail-org/… Talking about RAG, I recently wrote an article on a new approach that makes retrieval much more efficient by cutting corpus size by 40x, reducing tokens per query by 3x, and improving vector search relevance by 2.3x. The article is quoted below.

A 26-YEAR-OLD AI ENGINEER WHOSE BIRMINGHAM THESIS OUTPERFORMED GOOGLE SCHOLAR BY 50% JUST SHIPPED GRAPHIFY: ONE COMMAND TURNS ANY FOLDER INTO A CLAUDE CODE SECOND BRAIN Safi Shamsi built Graphify 48 hours after Karpathy posted his LLM wiki idea. It turns any folder, codebase, docs, PDFs, into a knowledge graph Claude reads instead of grepping. Up to 43x fewer tokens per query. The trick almost nobody is using yet: one flag exports the entire graph as a fully-linked Obsidian vault. Repo: /safishamsi/graphify Setup, end to end: 1. Install: uv tool install graphifyy (or pipx install graphifyy). Verify with graphify --version. 2. Install the skill in Claude Code: graphify claude install. This wires Graphify into Claude Code so you can call it as a skill. 3. Open the folder you want mapped in Claude Code. In the terminal: graphify . It extracts every concept and builds the graph in graphify-out/. 4. Export to Obsidian: graphify . --obsidian. Writes one note per concept, every relationship as a wikilink, every node linked back to its source. 5. Open the new vault in Obsidian (Manage vaults → Open folder as a vault), or drag it as a subfolder into your existing one. That's it. Your Claude Code instance now has a navigable map of the codebase that loads instantly instead of re-reading files every session. Full step-by-step build of the Claude + Obsidian second brain in the article below. Bookmark this

@adricazoorla Looks like a close up photo of a cut out in a marble pound cake. 🤔

alguien sabe dónde está esto ?

@TheDavidSJ 💯 and it’s the worst of the popular LLM’s for doing this.

Never great when Claude starts off with "I want to be straight with you"

@GrowingUpRetro Not doubt the guy is swimming in Drakkar. You can smell it through the freaking screen!

If the 80s were a person 👇🏻

@LangmanVince One word. Enron.

I still can't believe I voted for this RINO twice and defended him against radical leftists he now embraces for 10 years after he left office!

@Layemie001 This is the best video ever LMAO

Why do horses always decide to run as fast as possible when you’re new?