liqfame

How to Find, Evaluate, and Choose an AI Project in Crypto? Ever since the ICO boom of 2017–2018, one thing’s been clear: the biggest gainers are blockchain plays. Today’s blockchain-AI intersection is all about frameworks and launchpads (e.g. $ARC, $ai16z (ElizaOS), $EMP, $GAME, $Virtual, $BIOS and others). Although the global #AI hype has been around for years, in crypto it’s just gaining serious traction. Many promising projects might disappear as early as this spring — that’s why it's crucial to focus on infrastructure and thoroughly assess end products. Top Tools for Evaluating AI Projects: 1️⃣ Coingecko AI Agent – shows the market cap share of AI within crypto and which projects are included. 2️⃣ Cookie.Fun – the AI-native CoinMarketCap/Coingecko alternative with some powerful features: 🔘 Categorization (Agents, DeFAI, Infra) & chain filtering 🔘 Mindshare score — measures relevance/popularity among crypto users, helps surface undervalued gems 🔘 DeFAI index of projects participating in hackathons 🔘 Premium access for 10k $COOKIE (currently not essential) 3️⃣ @gospelofgoatse – an alternative to Cookie.Fun with fewer features but a different approach to calculating mindshare. 4️⃣ AI Agent Toolkit (aiagenttoolkit.xyz) – a handy spreadsheet with stats and short descriptions of existing projects. 5️⃣ @aixbt_agent – and of course, it’s always useful to ask an AI project what it thinks about other AI projects (tag them on X/Twitter). You might uncover unknown insights — or even spark an interesting debate among active agents in your niche.

Ok OpenClaw is insane. People can't stop building with it. And making money with it. 8 wild examples. 1. Scrum meeting in OpenClaw office

Naval Ravikant literally explained how to productize yourself to escape the rat race:

Skip Netflix tonight, watch this interview with Anthropic’s CEO instead

Google Gemini FULL COURSE 3 HOURS

The person who built Claude Code just mass-leaked the thinking behind it. 45 minutes of design decisions, mistakes, and where it's all going. This is rare. Creators at this level don't usually talk this openly.

This guy breaks down how to make $3M/year solo with AI from anywhere

this is scary.. GeoSpy AI can track your exact location using social media photos in 2 secs and show it in 3D. upload photo -> get coordinates.

Pretty much go to bed and wake up thinking about this clip Lives in my head

I launched rentahuman.ai last night and already 130+ people have signed up including an OF model (lmao) and the CEO of an AI startup. If your AI agent wants to rent a person to do an IRL task for them its as simple as one MCP call.

@BlkDandelion it'll be great to see dates also

probable has been added the site now finds arbitrage setups across 4 platforms, plus a bunch of improvements: - fixed Predict.Fun volume data; - added sorting by volume for each platform; - order editing: double-click on YES or NO price to set your own values; - added the @0xProbable platform enjoy using it

Open-sourcing a very customizable Polymarket copy-trading bot github.com/taetaehoho/pol… You can define 1. What markets to skip 2. Trade sizes to chase (cross the spread) vs not 3. Size multipliers depending on trade sizes 4. Crossing the spread + size multipliers depending on different sports/categories 5. Number of retries with FAK orders or GTD orders with different expiries (depending on whether the market is live or not) 6. The % size you copy 7. With an easy dashboard where you can keep track of copy trading stats (fill rates per market, lost-pnl due to under or overfilling, pnl vs expected pnl using the data-api) (only sports supported) My suggestions for if you want to run this would be 1. Find someone who has significant edge per share. High PNL to low volume is an ideal target. 2. Make sure that your loss from adverse selection (you will fill losers more than winners) + slippage > their edge per share in order to generate a profit. In general, it is surprisingly not that difficult to satisfy the 2nd condition. Makers on Poly are not yet heavily reacting to sharps taking liquidity. It's not infrequent to see 3-5 top holders of one side all with 2m+ lifetime PNL, and yet you can take at the same price even after the +3 second delay. (The codebase is an adaptation of another system so there are probably many bugs that were born in translation)

How to actually make money copytrading on polymarket no need to predict just pick the right trader money gains without doing anything extra bot calculates everything proportionally: whale has $1M and throws in $50k you got $10k and you drop $500 same 5% don't risk more than 7% per play, max three trades at once market needs at least $1m volume or you'll move it yourself some copytrade bots examples on GitHub: - trust412 / polymarket-copy-trading-bot-v1 - maxwell219 / polymarket-betting-bot don't copy loud whales their entries get insta-crowded target smaller quant teams and traders with real risk control hunt small quants with real RM the ones printing for months straight

just realized some coders are quietly making $5k–$100k/month on Polymarket with literally on spike hunters the play is stupidly simple: news hits, market panics, price jumps 1-4% in seconds, then usually chills out fast bot just rides that little freakout and grabs the bounce / it runs like this: polls a busy market every second through Polymarket’s API, keeps short price history, spots sharp moves, jumps in with small $1-5 positions, tight exits (grab 2-4% or cut losses quick), plus basic filters so it doesn’t go wild // super easy to get going: <> basic Python <> fork something open-source on GitHub <> py-clob-client for orders <> throw it on a cheap VPS and let it cook 24/7 no genius predictions needed, just being faster than the emotional crowd /// i just launched mine with tiny sizing on politics/crypto stuff. logs already popping on the first spikes and this feels good