moon2026 @FENDOU2024

@abindawood88 AI 画的？

想了3天3夜，怎么也想不明白 我怎么就把他做出来了 喜欢的点个赞，也许会开源出来

@Dark_start_over 这个nb啊 交流一下啊

HE IS GENIUS. THE GUY TURNED GOLD INTO A DIAMOND.+100 STARS FOR FEW HOURS. github: github.com/h100envy/gem-s… gemsearch just updated and now the spider doesnt just scan narratives. it catches emerging signals straight from the feed > spider pulls $TICKER mentions from posts, skipping majors and prices > catches solana addresses and only takes real ones that decode to a 32 byte key > finds new narratives: word pairs repeated by at least 3 different authors in 6 hours that arent in known topics > for each find it calculates growth: mentions in 6h vs previous 18h, author count and time of first signal > keeps top 10 tickers, 10 addresses and 5 phrases per pass so the grok limit goes only to the best finds > topics can be set with your own topics.json file not just built in ones > tickers and addresses are marked research only and dont enter the launch queue. protection from the launcher minting a copy of someone elses token the repo hit 100+ stars and 28 forks in a few hours. for a tool almost nobody knew about thats insane and a sign the project is serious before the spider found narratives. now it finds narratives, tickers, contracts and calculates their growth speed in real time. this isnt a chrome extension anymore this is a full radar

h100envy @h100envy

6 days ago

100+ STARS IN A FEW HOURS. I TOOK F*CKING GOLD AND TURNED IT INTO A F*CKING DIAMOND github: github.com/h100envy/gem-s… $GEMSEARCH doesnt just scan the feed anymore. it understands it. whats new: > spider catches tickers $TICKER straight from posts. majors like $BTC $SOL $USDC

55 15 143 44K 134

40 free sites crypto whales don't want you to know about who's buying, who's dumping, who's about to unlock. all free whale & wallet tracking 1.) arkm.com → see who owns any wallet 2.) app.nansen.ai → where smart money is moving 3.) debank.com → any EVM wallet's full portfolio 4.) app.cielo.finance → wallet alerts across chains 5.) whale-alert.io → every huge transfer, live 6.) lookonchain.com → whale moves explained in plain words 7.) kolscan.io → every KOL memecoin trade, realtime 8.) fomo.family → what top traders buy and why 9.) gmgn.ai → smart money and KOL tags on any solana coin 10.) solscan.io → solana explorer 11.) etherscan.io → ethereum explorer before you buy a memecoin 12.) rugcheck.xyz → insiders, LP and dev wallet on one page 13.) v2.bubblemaps.io → linked wallet clusters 14.) insightx.network → hidden connections between holders 15.) solsniffer.com → solana scam score 16.) tokensniffer.com → EVM scam score 17.) dexscreener.com → top traders, KOLs and holders tabs 18.) birdeye.so → solana charts and wallet PnL 19.) geckoterminal.com → any chain, any pool unlocks & insiders 20.) tokenomist.ai → unlock calendar 21.) defillama.com/unlocks → who gets tokens and when 22.) cryptorank.io → funding rounds and VC buy prices 23.) dropstab.com → vesting and investor lists leverage & liquidations 24.) coinglass.com/hyperliquid → hyperliquid whale positions, live 25.) app.hyperliquid.xyz/leaderboard → top perp traders 26.) coinglass.com liquidation heatmap → where the stops sit 27.) velo.xyz → funding, open interest, basis 28.) coinank.com → order flow and long/short ratio where the money actually goes 29.) defillama.com → TVL, fees, revenue 30.) tokenterminal.com → protocol earnings like a stock 31.) cryptofees.info → who earns the most fees today 32.) app.artemis.xyz → money flows between chains 33.) farside.co.uk → daily ETF flows 34.) dune.com → thousands of free dashboards 35.) messari.io → free research 36.) l2beat.com → L2 risk ratings 37.) ultrasound.money → ETH burn and supply 38.) mempool.space → bitcoin fees and blocks portfolio 39.) app.step.finance → solana portfolio in one place 40.) debank.com/stream → what wallets you follow are doing bookmark this. you'll need it next time you're about to ape

我开始理解高频交易的 Latency 对应的梯度层级。 1 秒 （s） =1000 毫秒 （ms） 1 毫秒（ms）=1000 微秒（μs） 1 微秒（μs）=1000 纳秒（ns） 毫秒级：软件在交易。 微秒级：机器在交易。 百纳秒级：专用硬件在交易。 个位数纳秒级：你已经在和电子信号、光速以及物理距离较劲。

keyboard_arrow_left Previous keyboard_arrow_right Next

matthias 0️⃣ @0xevolve

2 weeks ago

i see people want more HFT tricks so here we go probably most of you doing any serious hft know these simple ones but I think it's always good to see what hft optimizations actually looks like 01/ every action is sent over 3 ws channels at once why? just benchmark it, you

2 4 85 33K 179

كيف نستخدم الـBookmap باحترافية عالية؟🩻🔴 الـBookmap ببساطة هو أداة تخليك تشوف السوق من الداخل، مو بس حركة الشموع بعد ما تصير. يعرض لك السيولة المعلّقة في دفتر الأوامر، الصفقات اللي تم تنفيذها فعليًا، أماكن تمركز المشترين والبائعين، وكيف تتغير هذي السيولة لحظة بلحظة (حرفياً لحظة بلحظة). لذلك أفضل طريقة مو إنك تشوف خط أصفر قوي وتقول: خلاص هنا دعم أو مقاومة. الاحتراف فعلياً يبدأ لما تسأل: هل السيولة ما زالت موجودة لما السعر قرب منها؟ هل انضافت أو انسحبت؟ هل صار تنفيذ قوي عندها؟ وهل السعر استجاب لهذا التنفيذ، أو استوعبه وكأنه ما صار شيء؟ كذا تبدأ تقرأ الـBookmap صح. تابع معي هذا الثريد عبارة عن صور ، مع تمنياتي للجميع بالتوفيق والربح الوفير. 🤍📄👍🏽

د. سلطان بن جريس @SAJ20000

a week ago

صباح الخير يا أصدقاء .. أعتقد أنه وجب الارتقاء إلى Level أعلى. 🔴 الشارت وحده يريك ما حدث، لكن الـ Order Flow يبدأ يكشف لك كيف حدث، وأين كانت السيولة، ومن كان يضغط فعلاً. مع الوقت تكتشف أن السعر ليس مجرد شموع ودعوم ومقاومات، بل خلف كل حركة مزاد مستمر، أوامر معلقة، سيولة تُسحب،

2 10 76 37K 60

卧槽！给 TradingView 做付费指标的那家 LuxAlgo，把自家图表引擎开源了 Vela，495 Stars，Apache-2.0 开源，也是 LuxAlgo 自己在用的金融图表引擎。 以前做行情前端，要么买 TradingView 授权（贵、数据还受限），要么自己造图表轮子（难、bug 一堆）。 这个不一样，LuxAlgo 直接开源整套图表引擎，不是 TradingView 的壳，是自带 WebGL2 渲染器的 headless 核心，数据层 / 脚本引擎 / 渲染层全是插件化接口。 核心能力： 1️⃣ WebGL2 金融图表：实时 K 线高性能渲染，Canvas2D 自动兜底 2️⃣ 70+ 技术指标：MA、RSI、MACD、VWAP、SuperTrend、Volume Profile 都有 3️⃣ 22 种画图工具：趋势线、斐波那契、Pitchfork、波浪、Long / Short Position 直接用 4️⃣ 多图 Workspace：多图布局、Crosshair 同步、周期切换、对象树都做好了 5️⃣ 实时行情接入：Binance、Coinbase、Hyperliquid 开箱即用，也能换自己的数据源 6️⃣ 支持 Pine Script：配合 PineTS，可以把原来的 Pine 指标直接跑进 Vela 我觉得它真正有意思的地方，不是“能画 K 线”。 而是你拿到了一套可以继续长成自己交易终端的骨架。 行情、指标、策略、AI Agent、Broker，后面都能继续往上接。 想自己做 TradingView 风格量化终端、Crypto Dashboard 或 AI Trading 工作台的，这个很值得留着。 仓库：github.com/LuxAlgo/Vela

爱吃折耳根的Ace @_zheergen

a week ago

这下省事了！高频回测里最容易骗你的地方，不是策略，而是你默认自己每张单都能成交 hftbacktest，4.8K Stars，专门给 HFT / Market Making 做订单簿级回测。 以前做高频回测，大多数框架只管“信号出来就成交”，很少认真处理延迟、排队位置和盘口变化。 hftbacktest

9 9 41 121K 75

a quant built 17 trading strategies in Python and just put them all on a repo for free momentum, breakout, stat arb, options, Monte Carlo, it's all there

Ruuj @RuujSs

a week ago

Inventory risk is the foundation of market making. Pricing information is the next level. x.com/i/article/2094…

6 6 99 112K 242

短线交易的圣杯 - 12个被证明真正有用的网红指标/策略 建议保存收藏，美股币圈必备。

keyboard_arrow_left Previous keyboard_arrow_right Next

EMA 9 + Vwap + ADX strategy in the stock market

YOU'RE TRADING WRONG YOU WATCH ONE CHART AND STILL LOSE. CLAUDE WATCHED CRYPTO AND GOLD AT THE SAME TIME AND TURNED $1,000 INTO $9,423 IN 36 HOURS. Claude Opus 5.5 started with $1,000. Not $10,000. Not $100,000. One thousand dollars. Screen 01: crypto. Bitcoin, then Ethereum, then Solana, then XRP. Screen 02: metals. Gold and silver. Claude ran two full desks at once. Each one had its own order book and its own options radar. Here is what got me. At hour 18 the two screens sat in opposite modes. Solana was trending, so it shorted with the move. Gold was snapping back, so it bought the dip. Two markets. Two opposite playbooks. Same minute. Hour 1: $999.75. Flat. Hour 3: $1,080. Hour 6: $1,462. Hour 13: $1,982. $2,569. $2,607. $2,807. Hour 23: $4,165. Hour 28: $5,732. Hour 33: $7,039. The curve went vertical. Hour 36: $9,423. From $1,000. 9.4x. 31 trades. 71% win rate. 31 trades across two screens in 36 hours. Less than one an hour. It waited for every single one. $8,423 profit in a day and a half. From the price of one cheap weekend trip. Not once did a human switch screens, pick a side or click buy. Claude ran both desks alone. The full setup is below, free, and almost nobody runs it. Save it now, before the feed moves on.

INSIDER @x_insider4

a month ago

Your backtest is lying to you and you don't even know it. This is the $200K quant stack — rebuilt with Claude in one evening. Full code. Copy it tonight or keep trading noise. x.com/i/article/2094…

38 85 578 1.7M 3K

paste this prompt into claude code it will build you an Opus 5.5 + Jev trading bot it will probably make you a lot of money...

Roan @RohOnChain

3 weeks ago

Jev is the FASTEST AI model ever built for trading It makes calibrated buy/sell decisions in under 100 ms That is one real decision on every single block, 24/7 In this article I've shown EXACTLY how to build HFT trading system with Jev (from scratch) x.com/i/article/2100…

205 381 3K 5.2M 11K

If I had to start again with $1,000, this is the exact bot I'd build. I lost money twice when I started, once trading by hand and once running somebody else's bot. Step 1: Get the tools Download Claude desktop at claude.ai/download, open Claude Code. Use tradingview.com, the free plan gives 2 alerts and that's enough to start. Free account at tradingkit.com: My Account, copy the line of code, paste it into Claude Code and hit enter. Copy your authentication code from the same page and paste that in too. Step 2: Check the connection Prompt: "Hey there buddy, could you backtest a very simple strategy on the 1 hour, I just want to make sure the connection to my MCP server is working" Step 3: Build it the way every one of mine is built Prompt: "Build a trend following strategy on BTCUSDT 1 hour: 1 indicator for trend direction, 2 for confirmation, 1 volume or volatility filter to stay out of sideways markets, ATR bands for the stop loss and take profit. Size positions smaller in sideways markets and bigger in strong trends. Backtest with fees on, full history" Step 4: Set it hunting Prompt: "Loop this for the next 24 hours. Build new strategies, optimise them, backtest them, and log every one to a local HTML mission control with its equity curve, win rate and profit factor" Step 5: Incubate it Leave the survivors for 1 to 3 months, then prompt: "Re-backtest every strategy and split the results into the original backtest and the new data it has never seen" Green is the backtest, pink is live, if the pink holds up you've got something. Step 6: Move it to TradingView Ask Claude for the Pine Script, open the Pine Editor on tradingview.com, paste it, add to chart. Step 7: Connect your broker On your exchange create an API key and secret, read and trade only, no transfer permissions. Go to trigger.trade, pick your broker, paste the keys, copy the code it gives you. On TradingView add an alert, paste that code in the message box, paste the webhook URL, create. Step 8: Let Claude watch it The alerts MCP lets Claude read your results daily and switch alerts off when a strategy loses its edge. Real talk: 95% of strategies fail in incubation. I've tested thousands and trade 7. One bot is never enough, and $1,000 with 1 to 3% per position runs a few of them.

好了 这套系统 现在给老铁们开源。 我需要再忙其他的脚本了。 这个脚本经过长达2周的运行 调试 和升级 已经稳定无bug，加入了很多功能。 且里面包含了其他的系统。 辅助系统 除了网格，还有刷量 监控 为一体的系统 功能很丰富 也比较复杂。 用之前 一定要让ai帮你阅读一遍功能。 彻底理解清楚每个参数作用。 特别是网格系统。 我设计了很多的模式。对比早期版本 变化非常大。 后续 我会继续升级。 都会以这套系统的架构为基础进行升级。做新功能。 暂时先这样。 小白不要轻易尝试实盘使用。不对任何结果负责。DYOR （有任何疑问，你可以让ai帮你回答） github.com/cryptocj520/cr…

🚨 Someone built an AI that reads candlestick charts the way GPT reads English. Trained on 12 billion records from 45 exchanges. Outperforms every model by 93%. Live BTC demo. Free.

搞量化的还在一行行手撸技术指标？这个 Go 开源库 Indicator 了解一下。v2 版本一口气内置了 80 多种技术指标，MACD、RSI、布林带、KDJ 这些常用的一个不缺，回测框架也是现成的，实时策略照样跑得动。 我个人最看重这三点： 1️⃣ 指标和策略全部可配置，没有任何写死的预设值，参数想怎么调就怎么调 2️⃣ 数据流走 Go channel，习惯切片的也有 helper 转换函数兜底，不用来回折腾 3️⃣ 作者说 MCP 支持已经移除，但生态里还有 AI 联动的路子 嫌 Go 上手麻烦的，作者还同步维护了 TypeScript 版本。想动手的自己翻代码。 🔗 github.com/cinar/indicator 丑话说前头：这就是个工具，回测好看不代表实盘能赚，亏了别赖它。

新版本，睡觉💤很满意

🟢 $HYPE/USDT My Short Trade Setup Entry - 79.44 Leverage 50X Target - 77.08 SL - 80.43 #Tradingsignals #bbb26 #englot‌ Join our telegram group for more signals 👇👇t.me/+_dZn3Lf0GyxiM…

花大价钱搭量化系统的人，看到这个项目估计都要沉默了 这是一个可以部署在自己电脑或服务器上的 AI 量化交易开源平台，GitHub 已斩获 1.1w+star 加密货币、美股、A股、港股、外汇期货，一套系统统一研究和管理 可视化向导直接接入 OpenAI、Claude、DeepSeek 等大模型 让 AI 帮你分析市场、生成 Python 策略，策略写完直接回测 手续费、滑点、杠杆全部计算进去，调好参数先跑模拟盘，没问题再上实盘 还能整合行情、基本面、宏观数据和金融新闻，把重要信号通过邮件、短信推送到手机

阿西_出海 @axichuhai

a month ago

兄弟们，发现一个可以套利薅羊毛的比赛。 OKX 最近有个交易赛，现在还剩 3 天，前 800 名就可以瓜分 5 万 U 奖励。 我看了一下，只要交易额超过 200 刀就能上榜，最低也能分到 35 U。 这里解释一下什么叫交易额 就是你买和卖加起来总的钱。比如你买进 100 U，卖出 50 U，那么交易额就是 150 U。

keyboard_arrow_left Previous keyboard_arrow_right Next

38 0 4 35K 3