BigA

@Asteri_eth @threadreaderapp unroll

Karpathy found a way to reduce token consumption by 90% The problem is that the LLM re-reads the same files over and over again, loses context between documents, and provides less accurate answers as a result The solution is called Wiki Layer the LLM cleans, structures, and links all your data once, after which it never works with raw files again Three folders `raw/` for originals, `wiki/` for a clean knowledge base in Markdown, and files with rules for the agent Result up to 90% token savings on repeat queries, automatic links between documents, and a visual knowledge graph in Obsidian Everything stays on your local machine nothing goes to the cloud

@LORWEN108 @threadreaderapp unroll

A Stanford neuroscientist warns high cortisol wrecks memory, enlarges your fear center, and make your brain feel broken. If I wanted to fix it naturally, I'd do these 8 things every day: 1. Walk barefoot on grass for 5–7 minutes.

@edzitron This is the ChatGPT version of “don’t believe your lyin’ eyes”

??????

@mhp_guy @threadreaderapp unroll

Meet Phil. He's a retiree who didn't know how to code six months ago. Today he's closing $50,000 consulting contracts and building custom AI apps that clients gladly pay $3,500 for. He started by texting every contact in his phone. Over 1,000 people. And just asked to meet for coffee. With his first few meetings, he talked too much and tried to sell AI solutions instead of listening to their problems. But then he figured it out. Now, he just sits down, asks what's broken in their business, records the conversation on an AI pendant he wears, and plugs the transcript into Claude. Claude helps him build the solution, draft the contract, and even price the whole thing. One client is a chiropractor launching a new laser service. Phil charged $3,500 upfront and $1,500/month. Took him 12 hours to build the entire launch system. Another is an investment firm managing $1.4 billion. That contract? $50,000 for six months of part-time work. In this episode Phil: - Breaks down how he went from voice agents to $50K contracts in six months - Shows exactly what he built for each client and how long it took - Tells me why friends and family are the best first customers - Gives the framework he uses to stay motivated when nothing's working yet This one fired me up. Check it out.

@ridark_eth @threadreaderapp unroll

A 30-year-old solo developer decided to stop chasing trends and made $77,000 in a single month He built 35 different micro-SaaS startups while working completely on his own He didn't build complex AI agent teams He just paired a basic code editor with a single AI chat window Every single day, he follows a military-like routine: wakes up at 6 AM, hits the gym, and locks in For 4 to 6 hours straight, his phone is completely off. Zero social media, zero emails. Just pure deep work He doesn't even check bug reports or customer support in the morning to keep his focus clean Out of 35 startups he launched, 30 completely failed and made $0 But one single project (Trustm) now generates over $35,000/month alone His main secret? He ships features immediately instead of polishing them for months I recommend reading the article below

@0xFrogify @threadreaderapp unroll

This might be the biggest unfair advantage in content creation right now: Claude can now pull insights from over 135 million channels and 12 billion video Claude AI is now so good at faceless YouTube that normal people are building entire channels that look and perform like they have a full team behind them. No face on camera. No expensive tools. No guesswork. Just paste a few prompts and Claude handles the heavy lifting - viral ideas, scripts, titles, descriptions, even content strategy. This isn’t theory. This is happening right now in 2026 - and the gap is widening every single day. Most people will read this, feel that little rush of possibility… ...then scroll away and stay exactly where they are. The dangerous minority won’t. Follow me. I’m exposing every single AI-powered money printer I discover.

@insomnia_vip @threadreaderapp unroll

A 24-year-old generated an AI girl on his laptop and built a page pulling millions of views every month No camera No real person No photoshoots Just AI-generated models, fast edits and captions engineered to keep people watching One clip took 11 minutes to make It crossed 8,700,000 views in 5 days ChatGPT wrote the scripts Flux generated the model CapCut edited everything automatically Now the page runs 24/7 without him even touching the account Most people still think AI tools are “just for fun” Meanwhile anonymous creators are building entire audiences from people who don’t even exist

@Shelpid_WI3M @threadreaderapp unroll

@Bober_smart @threadreaderapp unroll

A 48-year-old girl from Britain decided to just make a joke, but in the end, she earned $2,245 in 4 hours With the help of Claude, she created a Chinese girl and started a stream Claude changed the appearance, background, voice, and even language in real time She started the stream on 3 platforms at once: YouTube, TikTok, and Kick Men liked the girl so much that they started donating > In 4 hours, she was donated $2,245 > The biggest donation was $400 The article contains 10 folders with tools for creation and workflow automation

@MidwesternDoc @threadreaderapp unroll

Bill Maher reveals what CEOs are privately concerned about with employees on Ozempic. MAHER: “I’ve been hearing this on the low for a while.” KARA SWISHER: "On the low. From who?" MAHER: “I’ll tell you after the show. Somebody [in Big Tech] you would know... He’s been hearing, and in his company also, they don’t like their employees on this stuff because—” KARA SWISHER: “Oh, they don’t work hard enough for the man?” MAHER: “They don’t. It takes away your craving for food and maybe substance abuse...” “...Also kind of for success, and sometimes for living and for being motivated. They get kind of logie.”🧵

@adiix_official @threadreaderapp unroll

watched a 20-year-old kid casually explain how he makes $10k/month while sleeping and i’m convinced 99% of people are missing this the entire setup: pick a YouTube channel that posts often. drop the link into one tool. plug in your TikTok, Instagram, YouTube Shorts. close the laptop. from that point on, AI does everything catches every new upload, slices the viral moments, captions them, blasts them across every platform. > 1M views ≈ $2,000 >zero editing, zero posting, zero stress > 10 min to set up, runs on autopilot forever you’re not making content. you’re owning a content factory. the full step-by-step is in the article below exact tools, exact stack, exact playbook Save it

JUST IN: Thomas Massie has officially filed a statement of candidacy, allowing him to raise money for a potential 2028 run.

@shedntcare_ @threadreaderapp unroll

🚨 Someone just built the open-source tool the internet has wanted for YEARS. It’s called ReClip — and it downloads videos from almost any website. 100% free. Self-hosted. No ads. No tracking. No nonsense. Paste any link from: → YouTube → TikTok → Instagram → Twitter/X → and 1000+ more sites Choose MP4 or MP3, pick your resolution, and download instantly. That’s it. No popups. No rate limits. No forced signups. No “premium” paywall halfway through. Just your machine, your files, your privacy. What makes it even better: ↳ Batch download multiple links at once ↳ Clean browser-based UI ↳ Lightweight setup anyone can self-host ↳ Built with simple HTML ↳ MIT licensed ↳ Open source forever And the craziest part? It went from 0 → 1.4K GitHub stars in just 9 days. Already crossed 239 forks. That kind of momentum says everything. For the last decade, people have been stuck using sketchy downloader websites packed with ads, trackers, and fake buttons. ReClip fixes that. A clean, private alternative built for people who care about ownership, speed, and freedom. No data harvesting. No subscriptions. No upsells. Just download what you need—on your own terms. Open source wins again. 👇

@Hi_entreprenuer @threadreaderapp unroll

DR. JAMES PENNEBAKER SPENT 30 YEARS STUDYING WHY SOME PEOPLE RECOVER FROM TRAUMA AND OTHERS DON'T. He found the same pattern in every person who healed. It had nothing to do with therapy or time. - THREAD 🧵

@testerlabor Please stop. Imagine is failing at every level and is being ratioed into oblivion. It is a dumpster fire and people are dropping Grok altogether because of it.

Grok Imagine was launched just 9 months ago and here are some of the things xAI already achieved: Imagine Apps Imagine on the web Quality Mode 720p videos up to 10 seconds Imagine API edit & longer videos Extend from Frame Imagine Agent Mode and a lot of Bug Fixes & Improvements

@achilleas_ghost @threadreaderapp unroll

The hormone your doctor never checks for: Cortisol When it's normal your body burns fat all day. When it's high, it stores fat around your belly and wrecks your other hormones. Here are 7 methods to reduce it naturally (backed by science):🧵 1. Sauna (5 nights/week)

@ridark_eth @threadreaderapp unroll

This girl turned a simple Claude-Canva connection into a $25,000/month money machine, bypassing years of design school. The income is incredible, but what really matters is the workflow itself. Every single image in this system is 100% generated using structured AI prompts. Agency-level skills are now completely democratized. You don't need a team, expensive software, or design experience to dominate the local business market. You only need one core system. The idea is terrifyingly simple: connect Claude to your free Canva account. Feed structured briefs into the AI to reverse-engineer professional templates. Swap out branding, aesthetics, and the entire copy using a single prompt sequence. One workflow. Endless niches. Absolute content leverage. Most people just scroll past AI and treat it like a chatbot. Smart players understand that combining LLMs with design APIs just captured maximum attention and revenue for them.

@xkaidus @threadreaderapp unroll

This guy cracked the code on AI virtual influencers using real-time face filters and now D2C brands pay him $2,000 per UGC video. He got tired of watching D2C brands burn $4,000 on a single creator who takes 2 weeks to deliver one angle, so he built a setup that runs hyperrealistic AI girls in real-time from his own webcam, generating viral content without actresses, studios, or makeup artists. His monthly revenue hit $89,000 last month from a network of 7 AI personas across TikTok and Instagram, while the average UGC creator caps at $6K juggling 4 brand deals. Here is the exact breakdown: → The hardware is the moat, but most people butcher the setup in the first frame. You need the face mesh locked at 60fps with zero artifacting → Persona comes first, and if you mess this up nothing saves it. Name, backstory, voice tone, niche before a single clip is shot → Face selection is not random. You A/B test features (eye spacing, jawline, hair contrast with face-framing highlights) because some faces convert better in 9:16 → You are picking who your audience trusts, not who looks cool. That is your targeting baked into bone structure → Real-time physics run before the script, and this is what kills the uncanny valley that destroys watch time in 2 seconds → The filter has to survive the strap of a tank top, the texture of a knit cardigan, the hair flick. → Batching is the move 96 percent skip: one performance, multiple personas, three platforms. → The system pushes 12 pieces of content before lunch, while traditional brands test 2 creators per week and wonder why their CPAs are stuck at $94 The economics are stupid: each video costs him $4 in compute, sells for $1,500 to $3,000, and takes 14 minutes to produce. That is a 37,500 percent margin, while UGC agencies pay creators $400 to $800 per clip and net $200 after revisions. One supplement brand generated 14 variants with 7 personas in 4 hours and found a winner in 36 hours without flying a creator to LA. They were previously paying $1,200 per UGC video and burning $6,000 per week on content that did not scale. Now they spend $210 for 14 variants and their CPA dropped from $89 to $27. The avatars hold real products. Warm window light on the persona, cold neon on the operator. Mouth shapes sync to consonants, not just vowels. Just a webcam, a tracked face, and the discipline to move enough that the filter never has a chance to break.

@SamTinnerholm @threadreaderapp unroll

Remember when we dropped the full Polymarket order book archive? 30M+ rows per hourly dump, raw parquet files, everyone loved it The problem: you had to download 500MB-1GB parquet files, write your own parser, reconstruct the book yourself from tick deltas That's over. You can now query historical order books directly from the API. 4 lines of Python. 🧵

@EricLDaugh Something’s not right. This is not a “deal has been reached assembly”. This seems like a “we have to make a bad decision” assembly.

🚨 JUST IN: President Trump is at the White House RIGHT NOW with his all-star team to discuss a deal with Iran, and host a conference call with multiple Gulf nations "LIKELY UNDERWAY" 🔥 JD Vance RUSHED back to the White House from Ohio, and Trump is "50/50" on launching bombs All eyes on 47 this weekend 🇺🇸