Fjärt

@plainionist Because Dario, Sam, and Jensen told me it does

If AI writes the code, and AI writes the tests, and AI reviews both ... how do you know your software actually works as expected? 🤔

@BetancesAndrew @getrichordietri @CollinRugg I think it bolsters his prominence in the algo regardless if its a post / reply etc. it means he's "relevant" and has things to say that gets people to engage. Lets ask the tards: @grok @gork do engaging replies help boost your account in feeds when you post?

@Poofarmer69 @getrichordietri @CollinRugg How does this work? There’s not way he’s making money off this right?

NEW: Police have released more info about the "Harry Dresden" break-in in Fairfield, CA, say both the homeowner & Nichols sustained head injuries. Police are praising the "actions of the homeowner" who raced home to confront 30-year-old Jason Nichols with a shovel. Authorities say that the homeowner's wife and child were inside when Nichols broke in. When Nichols was unable to break in through the front door, he broke in through a sliding glass door. "The homeowner's husband, who was away at the time, observed the suspect through a home security camera and immediately returned to the residence," said police. "As the suspect entered the home, the homeowner returned, armed himself with a shovel, and confronted Nichols." "A physical altercation ensued, during which both the homeowner and Nichols sustained head injuries." "We are grateful that the family is safe and commend our officers for their swift response in bringing this dangerous situation to a safe resolution." The woman and child were reportedly unharmed.

@getrichordietri @CollinRugg You're my hero

@Poofarmer69 @CollinRugg Are you sure?

@NWDisorder1111 @getrichordietri @CollinRugg This platforms lifeblood is ragebait. Their algorithms are designed to show you content you hate because you are more apt to click and interact when you are enraged

@Poofarmer69 @getrichordietri @CollinRugg Why doesn't Musk get rid of these worthless accounts. They are literally violating the TOS regarding engagement farming.

@NWDisorder1111 @getrichordietri @CollinRugg They are critical when presenting engagement numbers to advertisers…

@getrichordietri @CollinRugg Ragebait farmer alert

Everyone cheering the homeowner for beating a man with a shovel in front of his own kid? Y’all are wild. Jason Nichols broke in, sure, but the second the husband showed up swinging a weapon, it became attempted murder. Self-defense goes both ways. This isn’t bravery, it’s vigilante violence with a gardening tool. Free Jason Nichols, the real victim of overreaction. The praise for this guy is insane. Due process for Nichols

I worked out twice a day and was very fit. I’m a type 1 diabetic so I literally can’t eat bad food I’d be dead by now. I also can’t eat gluten or dairy due to other common comorbid autoimmune disorders, including hashimotos. Literally all I can eat is unprocessed meat, rice, salt and some spices and certain veggies and fruits. I hit my late twenties and started to have intense insulin resistance (at like 10% body fat, cardio every morning, weight training 3-5 days a week) and would gain weight anyway, went from 185 and lean to 210-220 while on that routine . And whenever Id stop I’d gain 30 lbs in 3-4 months, happened twice. It would take a year or two to get back down to 210-220 and refused to budge. My case is extreme but it highlights the problem that if your hormones are off it doesn’t matter how much you eat and workout, weight loss is a chemical reaction not a mechanical one. You can’t set fire to soaked wood. If you have too much free insulin, low testosterone, etc you can’t out work it, it disrupts the process. You can’t run enough of a deficit. People in the US have been fed processed poison foods for decades and a lot of ppls bodies suffer from it chemically, especially their guts. They should fix the food supply but instead they just make drugs to offset it. Seratonin? 95% is produced in your gut, but you’re given anti depressants. GLP1? Produced in your gut, but you’re given ozempic. It’s difficult to eat properly even when you try, even veggies are GMOs, sprayed with pesticides, grown in poor soils. It’s a much bigger problem than you think, and the corporations would love to have you be their zealot as they gaslight everyone about calories in calories out. FDA ensures food that keeps you sick to increase demand for the drugs that temporarily “fix” it.

@BrianPrescott48 Whatever happened to portion control & exercise? 🤔

$14 billion in Ozempic sales last year. Zero profit from Berberine—a natural alternative that's been working for 3,500 years. That's why they buried it. That's why your doctor never mentioned it. Time to take back control. Here's how: 1. Modern Greed… twitter.com/i/web/status/1…

@jlongster Isn’t this just RAG agent memory? Ex: docs.openclaw.ai/concepts/memor… #custom-memory-stores-vector-dbs-etc" target="_blank" rel="nofollow noopener">microsoft.github.io/autogen/stable…

regarding agent memory, I'm realizing: I never want anything loaded automatically. no loading yesterday's memories etc (AGENTS.md is different, not memory) I don't want chronological memories. I want topical memories grouped based on what I'm doing I want to be explicit about it generally, especially saving but also reading. some automatic reading based on me switching "into" a topic might make sense most thing I see don't really fit what I want

@om_patel5 They lose thousands on a lot of these subscriptions… each… their end goal was always enterprises paying millions to replace entire teams

saying "hello" to Claude on the Pro plan now costs 2% of your entire session usage one message. "hello, how are you?" that's it. this is why people are mass migrating to Codex right now because its literally impossible to reach limits anthropic needs to fix this before they lose the crazy amount of developers they just gained

they didn't need to import a new voting class back then, the left was sensible and more popular. now the left decided the best way to keep power is import illegals and let them vote, so they've been working people up into hysterics to go fight deportation to create martyrs so dipshits like you can post their photos and continue the brainwashing

@RubinReport Did any of their policy positions or stances bring about a similar incident? Did they rhetoric create chaos in some of American cities? Want to make sure I get this false equivalence right.

This is one of the most perfect things the internet has ever created.

@DrChrisCombs It’s one dog length

Evan you might want to ask for a refund

@johnanonle @JoshKale hahah exactly...

@JoshKale Wow what an improvement

Every website you’ve ever used is broken in a way you never noticed and it’s been this way for 30 years... A Midjourney engineer finally just fixed it. It’s called Pretext: A tiny library that lets websites lay out text the way magazines and newspapers do, with text flowing around images, wrapping into columns, and fitting perfectly into any shape, all at 120fps. This has been basically impossible on the web for 30 years. Every website you’ve ever used relies on the same clunky system from the 90s to figure out where text goes on screen. Pretext bypasses it entirely. 500x faster. The demos look like they shouldn’t be possible in a browser. Go look.

@JoshKale ans . Behind the looking good brother... smooth and seamless... can't wait to see single character columns at the cost for 4GB of ram per tab

@Noah_Kim99 @joeroganhq You should update your agent prompt to not include em dashes when tweeting, it’s a dead giveaway

@joeroganhq Bill Maher just demolished the reflexive “we’re paying for it” lie in real time. The ballroom is privately funded — yet some still treat every Trump project as taxpayer-funded by default.

Bill Maher: "I see so many people that are upset about [Trump] building a ballroom. I could[n’t] give a f*ckk." Mel Robbins: "Well, I give a f*ck because we’re paying for it." Bill Maher: "No, we’re not paying for it. It’s privately [funded]…"

That’s not exactly true… I read those docs then had claude build a doc skill that runs on a commit hook to keep docs fresh and up to date. makes sure I have memories matching glob patterns in rules/ for large sweeping requirements based languanges or parent directories etc. and then a CLAUDE.md in almost every folder containing symbols and basic usage. I keep my main CLAUDE.md below 500 lines. Usually just around 200. The rest gets lazy loaded based on the task at hand. I coupled it with Serena MCP for LSP retrieval. This drastically improved Claude’s ability to navigate the code base, and know what it’s working on without instantly filling up the context window. The context window is usually very low and only grows when it needs to. it also keeps my costs down significantly, like almost a 70% drop in token usage. Because each session doesn’t need to relearn anything, and has Serena preventing mass grepping bullshit. Every new session feels just as informed as the last. So reading the docs is extremely important especially when your project gets large.

I spent the weekend actually reading the Claude Code docs. It's a rabbit hole. CLAUDE.md files. MCP configs. Skills. Subagents. Hooks. Plugins. Agent Teams. You could spend more time configuring Claude Code than building software. All of it is productivity theatre. The only thing that actually matters: think first, then give it focused, relevant context.

@zostaff Hey can you show us results matching or close to a past real world event like 9/11, the 2008 financial crisis, etc?

2,400 MIROFISH AI AGENTS ARE TURNING MY $5K INTO $50K RIGHT NO there's an engine on GitHub called MiroFish - 22,000 stars, built by a 20-year-old Chinese student in 10 days got $4M in funding the next morning and right now my deposit on Polymarket is live and the equity curve is going up MiroFish does one thing: creates thousands of AI agents with unique memory and personality, drops them into a simulation and watches what happens i loaded 40 years of S&P 500 history into it and it generated 2,400 agents each with their own behavior: > panic sellers who dump on any red day > bulls who "buy every dip" institutions who "saw this in 2008" - retail who bought the top and is holding then i injected the current reality: rate 5.25%, inflation 3.2%, trump tariffs, war in the middle east, recession signals hit play: > 2,400 agents started arguing with each other,forming groups > opinion leaders pulled hundreds along > panic sellers triggered a cascade > bulls tried to buy the dip, 10 minutes later - consensus here's what MiroFish returned: Q1 closes negative - 79% of agents ATH before March - 4% of agents breaks below $6,400 - 34% of agents closes $6,500-$6,600 - 17% of agents quant models count numbers MiroFish counts people, my bot stacks both layers and finds where the crowd is wrong

@DaveShapi I can tell you haven’t actually tried coding full featured project with an ai agent yet

If your moat was "it took a while to code this up" then yeah, you're toast. And we haven't even gotten into the fully agentic economy yet. People saying "oh no we'll need MORE software engineers" are deluding themselves. Before long, you'll just talk to your phone like Tony Stark does to JARVIS or Picard talks to the ship computer. And it will figure out everything you need.

@nikmcfly69 Hey can you do an example POC introducing a real past world event so we can see if the agents actually respond like human beings did?

A Chinese student got $4M for an AI that simulates how thousands of people react to any event Problem: it was Chinese-only and required cloud APIs. I made it fully local + English Here's what MiroFish-Offline does and why it matters 🧵

@sachinyadav699 You’d think they’d show a few examples emulating past real world events to prove efficacy but nah…. Just trust me bro

$0 funding. a 20 year old spent 10 days building with AI. - now he can simulate 1000+ digital humans reacting to real world news. - we’re entering the era where one obsessed builder can create systems that used to require entire labs

so "moltbook: graphs edition"... cool very useful for society to see how text generators hallucinate based on different inputs for $30k in token usage per experiment. interesting how they haven't simulated a past real world event or two to show us how close the hallucinating text generators behave to the way people actually do to prove efficacy first

Mind blown: A Chinese quant college student builds an AI swarm engine in 10 days flat, explodes GitHub with 13,000+ stars, and scores $4,000,000 in funding! Introducing MiroFish is the multi-agent simulator that's revolutionizing predictions for trading, PR, and more. What is MiroFish? It's a digital sandbox where thousands of AI agents with individual memories and behaviors interact like a real society. Feed it any scenario (news leak, policy change, or even a classic novel's missing ending), and it simulates crowd reactions, debates, and outcomes to forecast real-world events. The Creator's Story: > In late 2025, fourth-year student Guo Hanjiang coded the core using AI assistants. > It went viral overnight, landing him 30m Yuan (~$4m) from Shanda Group. > He ditched the dorm, started a company, and now leads the charge. Key Applications: .Trading: Input financial news or reports, watch simulated market panics and price swings for predictive insights. .PR Testing: Companies/Politics run draft statements to spot backlash and refine messaging. .Creative Experiments: Loaded a lost-ending Chinese novel, agents role-played characters and generated a logical finale. .Easy setup: Deploy via Docker in minutes with any LLM API key. Pro tip: Simulate something wild like Elon Musk tweeting about Dogecoin 2.0 and spawn agent traders, influencers, and investors, generate real-time video clips of the frenzy to test moonshots or crashes risk-free. Traders are already winning big: Check this one on Polymarket - $120,000+ net profits from spot on SPX 500 bets, powered by MiroFish sims on historical data. His profile: polymarket.com/profile/%40moi… For effortless gains, try Kreo copy trading: Auto-mirror pros like him and ride their edges. Try here: @join" target="_blank" rel="nofollow noopener">kreo.app/@join Add his wallet: [0x17559efac103ac7f361be37ec0b93888d4c55aac] to [t.me/KreoPolyBot?st…] and start track/copy him. Repo: github.com/666ghj/MiroFish