slash1s

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

THE MAN WHO WROTE THE TEXTBOOK THE ENTIRE AI INDUSTRY LEARNED FROM WALKED ON STAGE AND SAID WE ARE BUILDING THE WHOLE THING WRONG. NOT THE MODELS, THE GOAL ITSELF. MAKE A MACHINE BETTER AT THE WRONG OBJECTIVE AND YOU DON'T GET A SMARTER TOOL -- YOU GET SOMETHING YOU CAN'T TURN OFF. 17 minutes from Stuart Russell -- Berkeley professor and author of the book that trained a generation of AI researchers. -> His warning: for 70 years we built AI to chase whatever goal we type in. That was always a bomb waiting for enough intelligence. 01:25 -- The gorilla problem. We once created something smarter than us by accident. Ask the gorillas how that turned out. 04:35 -- The coffee problem. A robot told only to "Fetch coffee" learns not to be switched off -- because dead robots make no coffee. 08:19 -- His fix flips it. The machine must stay unsure what you really want, so it keeps asking instead of steamrolling you. 10:43 -- The catch he admits out loud: we behave badly, our goals are messy, and the machine learns from that mess too. And now everyone is racing to hand real power to agents that were built on the exact model he says is broken. More capable, same blind spot. You thought safer AI meant smarter AI. This is the man who wrote the book saying more intelligence on the wrong goal is the danger, not the cure. Save this. It's the clearest 17 minutes on why alignment matters ↓

@0xwhrrari insane info, thx for sharing

Anthropic just leaked their 2026 agent roadmap In a 22-minute breakdown from the Claude team They walk through tools, memory, observability, and the parts most builders are still 12 months behind on The last 3 minutes alone are worth the watch Bookmark and watch it Then read the article below

@undefinedKi and it’s look fantastic haha

This is the single highest-leverage AI setup you can build, and I don't get why more people aren't doing it Andrej Karpathy posted the idea and it hit millions of views: stop using AI to write code, use it to build a wiki of everything you know. What people actually use it for: > Research that compounds. Drop in every article, paper, and transcript on a topic. Ask across all of it at once. Setup: a raw folder for sources, a wiki folder for the pages Claude writes, and one instruction to ingest. > Never repeating yourself. Your business context, your projects, your constraints all live in the vault. Every new chat starts with Claude already knowing them. > Decisions with a paper trail. Log what you tried, what worked, what burned, and why. Six months later, ask why you made a call and get the actual reasoning back. > Finding what you already knew. That article you saved, that idea you had, that problem you already solved. It surfaces on question, not on memory. > Writing from your own material. Ask it to draft something and it pulls from your notes and your voice, not the internet's average. Everything stays in plain markdown, so it's yours and it ports to any model. Five minutes to set up, and you never start from a blank chat again. Bookmark this

@ridark_eth yeah it was old way..

Paying developers by the hour is the old way. $15,000 in one month, and the team was Claude agents. Not one agent. A pipeline. One plans the work, one writes the code, one tests it, one checks for breaks. Each agent runs its piece. They share context. Nothing waits on a human in the middle. The billing stack matters. Basic tasks route to cheaper models. Complex tasks go to the capable ones. That gap cuts API costs significantly. RuFlow handles the routing automatically. 60 agents can run at once. It has 14,100 stars on GitHub and costs nothing to run. The developer pairs it with Obsidian for notes and task structure. Prompts go in, output comes out, the pipeline handles the rest. Month one on this setup: $15,000. That number is from one person's workflow, not a case study. The Claude subscription stays. The $200 platform fees go. The difference is who controls the stack. Open source means you fork it, adjust it, run it on your own infra. No vendor owns the ceiling.

@0xCristal I love listening him

Boris Cherny, creator of Claude Code at Anthropic, revealed why he believes coding is solved and hasn't written a single line of code in 2026: - 00:55 - Most of the room at Sequoia's AI Ascent already uses Claude Code daily - 03:35 - Coding evolved from typeahead autocomplete to full autonomous agents in under two years - 05:07 - "Coding is solved" for the kind of code he writes. He now ships dozens of PRs a day from his phone - 10:26 - The SaaS model as we know it is heading toward collapse because anyone can build their own tools now This is what the person who literally built Claude Code says behind closed doors at a Sequoia partner event. Not a hype thread. Not a demo. His actual workflow and predictions for the next 12 months. Watch the full 25 min before your competitors do.

@gippp69 that’s a lot start ngl

THIS F**KING 7,300 STAR REPO TURNS CLAUDE FABLE 5, CODEX, AND CURSOR INTO ONE AI TEAM THAT KEEPS WORKING WHILE YOU SLEEP 00:01 he opens Omnigent, an open source layer that puts multiple coding agents inside one workflow while sandboxes and policies control exactly what they can touch. setup takes minutes. Fable acts as the supervisor, cheaper models handle execution, and premium tokens are reserved for decisions that actually need judgment. each task gets split across separate agents and git worktrees, then a fresh model reviews the final diff before anything can merge. a $5 daily cap, 150 line approval limit, automated signoff tests, and a 95% pass rate keep the whole team from running wild. bookmark this. one repo, four agents, and roughly 15 to 20% of the usual token cost.

@sopersone @hyper_stack yep I already tried it

I've long believed: a strong prop firm isn't one that pulls you onto a new platform, but one that builds into the trading environment you already use. @hyper_stack does exactly that it works on top of Hyperliquid's existing infrastructure, without forcing users to switch anywhere. In my view, this is one of the most logical ways to strengthen the $HYPE ecosystem from within.

THE GUY WHO RAISED $55.5M FOR HIS AI COMPANY AND CO-CREATED STANFORD'S DEEP LEARNING CLASS WITH ANDREW NG -- JUST GAVE A 90 MINUTE MASTERCLASS ON WHY FINE TUNING IS A TRAP AND HOW REAL TEAMS BUILD AI AGENTS IN 2026 Kian Katanforoosh -- CEO of Workera. Clients include Samsung, Accenture, Siemens, and the US Air Force. Taught AI to 4 million people and won Stanford's highest teaching award. His pitch: the difference between a toy and a product isn't the model -> it's the architecture around it. > The jagged frontier: AI helps in some tasks and quietly ruins others. BCG consultants who trusted it outside its lane "Fell asleep at the wheel". > Two survival modes: Centaurs (Delegate big blocks) and Cyborgs (Rapid back and forth). > The fine-tuning trap: By the time you finish tuning on your Slack history, the next base model already beats you. Ross Lazerowitz tried it -- got a lazy coworker who kept saying "I'll do it tomorrow" instead of actual work. > Chaining beats one giant prompt: Extract -> Plan -> Write. Every step debuggable and every step testable. > RAG isn't optional: Vector DB + HyDE (generate a hypothetical answer, then search for it). This is what stops your model from hallucinating in production. > MCP is the shift: Anthropic's protocol so agents discover what APIs can do without you hard coding every integration. > The paradigm change: Engineers used to write deterministic code. Now you design "fuzzy" systems, think like a manager, decompose work into roles, keep a human in the loop for the messy parts. Stop trying to build a bigger model. Learn to orchestrate the ones we already have. Watch it, then Bookmark ↓

@refudgio looks unreal haha

OBSIDIAN'S CEO AND ANDREJ KARPATHY JUST QUIETLY KILLED THE SECOND BRAIN INDUSTRY it started as a single gist. 21 million people read it before anyone noticed what it actually meant the idea: stop treating your notes like a diary and start treating them like a codebase. Obsidian is the editor. Claude Code is the engineer working inside it. your notes are the thing being built three commands run the whole thing: one drops in anything, a video, a PDF, a random thought and splits it into small linked pieces that plug straight into what you already have written one lets you ask a question and get an answer pulled from your own words, your own pages, not a guess from a model that's never met you one runs quietly every week, hunts down the notes that contradict each other, deletes what's gone stale, and reconnects the orphaned pages nobody linked back then the Obsidian CEO shipped something that wasn't an "AI button" - five small skill files that teach Claude to speak Obsidian's own language natively. no broken links, no messy imports that repo hit 41,000 stars in weeks Karpathy pointed the whole thing at 100 of his own articles, 400,000 words, and just let it run. it links itself. it audits itself. it never sleeps no database. no subscription. just folders of text and an agent that does the one job humans always quit halfway through: keeping the filing system honest most vaults are graveyards. this one reads itself before you wake up

@shmidtqq @prophetmarketai btw looks beautiful

i just found @prophetmarketai and i wasn't expecting it to work like this. most prediction markets need a stranger to take the other side of you. - prophet has an AI take the other side instead. - a consensus of 5 models prices every market and becomes your counterparty. - so your position fills instantly. no empty book, no waiting for a match. - and anyone can spin up a market on almost any yes/no question in under 30 seconds. - you're not stuck with whatever the platform decided to list. so if there's a niche you actually read better than a general model, you can create the market yourself and back your read. i went through it and figured it was worth sharing.

@antpalkin haha yeah I feel it too rn..

Goldman quants realizing the Nobel laureate who runs Google DeepMind just gave their $650,000 seats a few short years before AGI runs the whole desk from a chat box

@monokern @Polymarket @PolymarketTrade good luck bro

One of the biggest World Cup matches so far Mbappe vs Yamal The face of this generation against the face of the next one I'm taking France 🇫🇷 Let's see if the dictator reminds everyone who's still in charge

@martynov014 it’s almost here 👀👀

The World Cup final is tonight and this trader has $145,000 riding on France Not spread around. Not hedged. One team, all in - and his track record says don't laugh You can copytrade him here betonstride.com/r/profit He only trades soccer - literally 100% of his book 22 bets all time, every single closed position green, 13 for 13 +$200K profit, +35% ROI, and Stride's edge metric at +18% - he wins nearly a fifth more often than his entry prices imply And France isn't a new crush, it's his whole tournament: France -1.5 spread - +$118K, his biggest win France -1.5 again - +$75K Norway to eliminate Brazil at 31 cents - +222% Match totals stacked green underneath Tonight's book: $131.7K on France to win the final, in at 41.6 cents $13.6K on France to lift the trophy Both slightly red right now - he hasn't moved a dollar A specialist, one read, everything on it, hours before kickoff That's either the best post on this app tomorrow or a very expensive lesson - and his full track record is public either way One honest note - 22 bets is a young sample, his max drawdown is -37%, and no streak survives forever Copying rides his conviction, swings included Stride is copy trading for sports prediction markets only The feed filters out the fakes and shows what you would have made copying, after slippage One tap before kickoff and tonight you're on his side of the final

@sopersone let’s go france

This STRIDE trader just placed a $758,624 bet on the World Cup semifinal match His prediction: 🇫🇷 France vs. Spain 🇪🇸: Team to Advance at 60 cents Such a big game and such a massive bet caught my attention too - so I also placed a bet haha I put in $194 with a potential payout of $335 Let's see what happens next, live in two hours France, come on, do it! 🇫🇷⚽

A $800 USED RTX 3090 JUST HUMILIATED NVIDIA'S $4,000 DGX SPARK ON LOCAL LLM SPEED Someone benchmarked 19 GPUs and found that the "AI-ready" flagships get destroyed by used gaming cards. VRAM decides what runs. Bandwidth decides what actually flies. Same 13B model, only the memory bus changed: 936 GB/s -> 19ms per token. 256 GB/s -> 70ms per token. 3.7x speed difference from bandwidth alone. The traps: > NVIDIA DGX Spark, $4,000, 128GB memory but only 273 GB/s bandwidth. > AMD Ryzen AI Max+ 395 mini PC, up to $4,000, 96GB unified memory at 256 GB/s. The actual value picks: > Used RTX 3090 -> $600 to $1,050 -> 24GB at 936 GB/s. > RX 7900 XTX -> $800 to $1,000 -> 24GB at 960 GB/s. The cores were never the bottleneck. The memory bus was, and the marketing has been lying about which spec to look at. The full 19-GPU ranking is in the article. Save this before you overpay for a "supercomputer" that cannot feed itself ↓

@antpalkin he is teal legend bro

Demis Hassabis, Nobel Prize winner and CEO of Google DeepMind, just called it: AGI is a few years away. His exact words yesterday: "we are standing in the foothills of the singularity" Everyone's sharing the safety parts. Almost everyone skipped the one line that matters for your money: "10x the Industrial Revolution at 10x the speed" Run that math. The last Industrial Revolution minted Rockefeller - $400 billion in today's money, the richest man who ever lived. It took him 40 years of railroads and oil fields. Hassabis is telling you the next one compresses that into a few years, and it's made of software. And new money always shows up in markets first. No factories to build, no permits, no supply chains - just data in, positions out. The "recursively self-improving systems" he says the world isn't ready for? Early versions already trade against you every day. That gap between people who run them and people who don't is exactly what he means by 10x speed. He wrote this essay to ask for guardrails before the wave hits. Read it twice and you'll notice what it really is: the most credible man in AI telling you the wave is already visible from shore. Bookmark this. In a few years, "I read Hassabis in the foothills" will sound like "I bought Bitcoin in 2013" - and you'll want proof you were here. I write about this frontier every week - AI agents and what they're doing to markets. Follow me, stay in the AGI foothills.

@gippp69 this is for real dangerous systems..

SOMEONE GAVE CLAUDE FABLE 5 A CONTRACT, A $5 DAILY BUDGET, AND A MANAGER. THIS F**KING DANGEROUS SYSTEM CAN CLEAR A WEEK OF BACKLOG WHILE YOU SLEEP. 00:03 he opens Microsoft’s 4,900 star Agent Governance Toolkit, built for policy enforcement, isolated execution, identity controls, and protection against all 10 OWASP agentic risks. setup starts with 3 files. CONTRACT MD sets the limits, boundaries MD defines what the agent can touch, and signoff SH runs every test before anything ships. four roles split the shift. one model reads the logs, Fable picks the highest value task, another writes the code, and a fresh Fable reviews the final diff. a $5 cap stops runaway sessions, commits above 150 lines need approval, and autonomy unlocks only after 20 runs at a 95% pass rate. one failed run sends the agent back to probation. the repo adds the guardrails, but the contract is what turns Claude from a chatbot into an employee.