Neo @code_87k

刚在苹果手机上用美区PayPal绑招商VISA信用卡，订阅ChatGPT太丝滑了，秒开，无需任何身份认证！

keyboard_arrow_left Previous keyboard_arrow_right Next

兄弟们，我一直有个判断：OCR 这种活，早晚会被多模态大模型给吃掉。 这周看到百度发的 PP-OCRv6，我改主意了。 一个 1.5MB 的模型，能直接塞进浏览器里跑，单图最快 97 毫秒就能出结果，逐字识别的准确率还反超了 GPT-5.5、Gemini-3.1-Pro 和 235B 参数的 Qwen3-VL，我有点震惊了！ 对做产品的人来说，这比「又一个 SOTA」重要得多。我有测试，先往下看 👇

@sven_ai 总是说的那么好

输入任意 App名称，自动抓取AppStore用户评价，用 LLM 做数据分析。 把评论变成产品经理能用的信息。 预设全球各国免费版和付费版Top10 App数据。 方便研究学习，代码已开源，见评论区

keyboard_arrow_left Previous keyboard_arrow_right Next

前身是阿里集团内部官方 AI 代码审查助手，过去两年服务了数万开发者，识别了数百万个代码缺陷。 现在它开源了：Open Code Review。 它不是让 AI “随便看一眼 diff”，而是把 AI Code Review 工程化。 它能读取 Git diff，结合上下文分析代码变更，生成带行级定位的结构化审查意见。 对开发者：提 PR 前先自查，少返工。 对 Reviewer：少看低级问题，把精力留给架构和业务逻辑。 对团队：可以沉淀规则，接入 GitHub Actions / GitLab CI。 一句话： 它不是取代人类 Review，而是让团队代码审查更省力、更稳定、更有价值。 GitHub： github.com/alibaba/open-c…

this is f*cking gold How to build your first AI agent (Full guide) if I had this a year ago, I would've shipped my first agent in a day instead of 2 weeks in the right hands, this changes everything:

Khairallah AL-Awady @eng_khairallah1

6 days ago

x.com/i/article/2065…

30 144 863 1.9M 4K

@Oscarweb0 hi

A Mac Mini sitting on a desk made $19,522 in 90 days. Nobody touched it. Three times a week at 6am, the pipeline triggers. Script. Voice. Video. Upload. Forty minutes later, three videos are scheduled. No face. No editor. No team. Cost per video: $1.50. n8n orchestrates every workflow. Google Trends finds topics. Claude writes scripts and SEO. ElevenLabs voices it. Pictory renders video. YouTube API uploads on schedule. One Mac Mini. Four workflows. Pays for itself in four months. Costs $10/month after that. Nine videos a week. $8,000/month from ads alone. Sponsorships from month three. Affiliate links passive in the background. Two channels: $16,000/month. Three: $24,000/month. Copy the workflows. Change the voice ID. One hour of work.

Superior @andreysuperior

2 weeks ago

x.com/i/article/2063…

12 12 120 126K 305

散户投研的外挂来了 Serenity @aleabitoreddit 白毛股神爆火，核心不是追龙头，而是找供应链瓶颈 我把 GitHub 上能找到的 Serenity 风格 Skill 全部挖了一遍，整理出这 10 个仓库 装进 Claude / Cursor / Codex 后，你的 Agent 就能直接套用这套框架做产业链拆解、卡点定位、标的初筛和 thesis 验证 1/ 核心方法论蒸馏版 完整复刻供应链瓶颈分析流程 热点拆链 → 找卡点 → 筛公司 → 核证据 → 排优先级 github.com/muxuuu/serenit… 2/ 推文 archive + Skill 版 基于 Serenity 公开推文和观点整理，带 ticker 知识库和周期跟踪，适合持续跟进她的最新焦点 github.com/yan-labs/seren… 3/ 多版本辩论融合包 融合 Serenity 和其他交易 / 研究思路，让 AI 从不同角度碰撞同一标的，适合做多视角 thesis review github.com/lanfuli/aleabi… 4/ Alpha 假设生成版 适合把市场新闻转成可验证的 alpha 假设 新闻 → 需求变化 → 财报映射 → 小盘弹性 → 验证路径 github.com/haskaomni/sere… 5/ 0xAgata 粉丝精炼版 轻量、直接、好上手，适合快速体验 Serenity 式供应链瓶颈研究 github.com/0xagata-prog/s… 6/ W-Y-P 完整框架版 基于大量公开内容蒸馏出的投资框架 Skill，通用性强，适合做系统化产业链研究 github.com/W-Y-P/Serenity… 7/ LeslieYeo 深度资料版 包含 Serenity 思维框架、表达风格、投资决策、外部评价、时间线等资料，更像完整 persona + research framework github.com/leslieyeo/sere… 8/ ZadAnthony 中文 Claude Skill 版 面向 Claude Code 的中文 Skill，聚焦 Serenity 供应链卡点逆向投资方法论，中文用户上手很快 github.com/ZadAnthony/ser… 9/ A股卡脖子选股版 把 Serenity 的 chokepoint / bottleneck 思路迁移到 A股场景，更适合做国产替代、设备材料、上游卡点筛选 github.com/fadewalk/seren… 10/ AI 基建 / 半导体供应链分析版 面向 Claude Code 和 Cursor 的 AI infrastructure & semiconductor supply chain analysis skill，适合拆 AI 硬件、光模块、半导体设备材料等链条 github.com/zongmin-yu/ser… Serenity 这套方法的价值，不是让 AI 直接告诉你买什么 而是逼你用更机构化的方式去问问题 这个环节是不是刚需 扩产是不是很难 认证周期是不是很长 供应商是不是很少 下游巨头是不是绕不开 财报里有没有验证 市场是不是还没 price in 装上这些 Skill，相当于给 Claude / Cursor / Codex 配了一个 24h 在线的供应链研究员 最后提醒 股市有风险，以上内容仅供学习和研究参考，不是投资建议 Skill 是研究工具，不是印钞机 仓库代码和具体标的都请自行 DYOR

@X_FINALBOSS X

Woke up 11 AM $7000+ made while I slept selling ebooks Each sale was $300 and the ebooks were generated by Claude AI + Perplexity + Nano Banana here's how I did it in 17 steps : 1. I created a new X account , posted anywhere between 1-5 posts per day ( AI generated ) 2. posts include : 1. engagement farming 2. case studies 3. stories 4. educational 5. threads 6. Graphs (cycle) 3. connected with tweethunter.io SaaS to X to allow auto DM 4. basically whenever someone comments or likes on any of the posts , it sends a checkout link to buy an ebook 5. I made 2 ebooks at 2 different price points 6. one is $300 and the other is a bundle of 3 ebooks at $475 7. they were generated with Claude AI within a few minutes 8. make sure that it looks great and there are many images and diagrams inside 9. you're probably thinking that it's overpriced but they also get few videos + discord community access + with good marketing , you can sell at any price point 10. I'm not revealing the account or niche just because I want to scale up to $30K per month first without any competition 11. but any MMO (make money online) niche works very well. choose one you enjoy or have experience with 12. if you don't have any experience, do affiliate. find top 5 whop courses and sent traffic to them and get 30% of the cut 13. this entire business costs $38 per month ($30 for tweethunter and $8 for X premium) 14. to get your first viral post , you will need to have very good copy AND get retweet boost from another account with many followers ( you can message me and I can do this for you ) 15. I did a launch which is why I had explosive sales within the first hour. basically built up audience and announced launch 5 days before launch. 16. hopefully this account will hit $100K within 30 days from now but I'll keep you guys updated 17. for those who cannot afford the $300 or $475 there is a downsell to $40 per month lower ticket. Also you can track IP so if they're from 3rd world country, auto redirect to low ticket MRR. If you want the full details + 2 hour free course video explaining how this business works comment "X' and I'll send it to you. **must be following + Retweet

A guy built a system of 7 Claude agents on his MacBook. No assistant. No sales team. No office. Every day it scans Google Maps across 3 cities, finds small businesses with no website or one from 2014, builds a landing page mockup, renders a 10-second video of it, and sends a personalized cold message — before he wakes up. 47 clients a month. $400 each. $18,800/month. $480 in API costs. Traditional web agencies run 8-person teams for the same order flow. He runs it alone from a MacBook and an iPhone. When a positive reply comes in while he's in a taxi, his Mobile agent books the Zoom call. He taps "approve" and joins 10 minutes later. The only time the system wakes him is when a deal breaks $3,000 or the reply rate drops below 12%. Everything else runs without him. Here's the complete playbook for building such $10K/month passive income machine with AI ↓

Rahul @sairahul1

2 weeks ago

x.com/i/article/2063…

15 45 348 769K 1K

I think the challenge is that everyone can now build apps But 1) almost nobody has distribution (like an audience), or 2) the money to pay for distribution (ads or UGC), or 3) the creative genius to get distribution for free (classically called guerilla marketing)

@

57 years ago

0 0 0 0 0

CURSOR está regalando 1 año de Pro GRATIS solo necesitas conseguir un correo de estudiante, tuyo o de alguien que conozcas lo que incluye: → 12 meses de Cursor Pro (valor $240) → acceso completo a GPT, Claude y Gemini → sin tarjeta tras la verificación cómo conseguirlo: 1. entra en cursor.com/students 2. regístrate con un email universitario 3. pasa la verificación de estudiante en algunos países es con un email .edu, pero cada país tiene su propio sistema de verificación estudiantil mientras todos pagan $20/mes, tú puedes programar proyectos enteros GRATIS durante un año aprovecha antes de que termine el plazo

Nico @nicos_ai

2 weeks ago

🚨OpenAI está regalando $1.200 GRATIS para usar Codex Solo necesitas un repo público en GitHub, rellenar un formulario y te dan 6 meses de ChatGPT Pro + Codex Y casi nadie está hablando de esto, seguramente porque no quieren que se entere todo el mundo Todavía estás a tiempo

462 130 1K 1.4M 2K

终于有人把枯燥的监控面板，做成了一张实时服务地图。 Maple 是一个开源项目，可以把后端系统里的 telemetry 数据，变成可交互、会动的服务依赖图。 对做微服务和分布式系统的人来说，这个痛点太熟了： 日志一堆 指标一堆 trace 一堆 出问题时还是不知道到底卡在哪个服务、哪条查询、哪个队列 Maple 的思路是直接把数据流画出来： 实时服务依赖 API / 数据库 / Kafka 交互关系 节点级 p99 延迟、错误率、性能指标 重查询定位 OpenTelemetry 接入 还带 MCP code server，方便 AI Agent 帮你审查系统 重点不是“图好看”。 而是排查线上问题时，你不用再在一堆冷冰冰的面板里猜。 你能直接看到数据怎么流，瓶颈在哪里，哪个服务正在拖垮系统。 对后端团队来说，这类可视化 observability 工具会越来越重要。

看到这个，我默默取消了剪映订阅… Clypra，GitHub 上快速崛起的开源视频编辑器。 用 Tauri + React + TypeScript 打造，专注把 CapCut / 剪映 Pro 的付费高级功能全部本地免费化。 跨平台（macOS/Windows/Linux），FFmpeg 驱动，原生性能。 以前： 剪视频要么忍水印、要么每月交订阅、要么上传云端赌隐私。CapCut 越“聪明”，本地能力越弱，创作者被平台死死锁住。 现在： Clypra 把专业时间线、多轨编辑、实时波形、胶片预览、精确裁剪全搬到本地。 零联网、零账号、零付费、零水印，彻底把控制权还给创作者。 核心亮点： - 本地专业时间线：多轨 + 标尺 + 播放头 + 帧精确，像 NLE 一样流畅，却轻量桌面级 - 免费复刻 CapCut Pro：高端文字、模板、特效快速跟进，再也不用为 Pro 按钮掏钱 - 纯本地 FFmpeg 处理：导出飞快，隐私安全，素材永远只在你硬盘里 - Tauri 原生性能 + 现代 React UI：启动快、占用低、深色模式丝滑 - 完全开源 MIT：想加功能直接 PR，社区一起把免费版卷到极致 - 跨平台 + 移动端已在规划：今天桌面，明天手机也能自由剪 内容创作者、独立开发者、AI 视频玩家，建议 Star 收藏。

我被蒸馏了，@golden_pan1 做一个美股投研网站把我过往的推文内容都抓取过去总结成了一个知识库，可以关注下。这里介绍下golden，美国佐治亚理工的本科西北大学硕士，在Meta最核心的AI广告部门工作已经近七年时间，躲过Meta多轮裁员优化（可见能力和重要性），平时也做投资，然后就Vibe coding做了一个美股投研网站，有兴趣的可以去看看

Golden Pan @golden_pan1

2 weeks ago

@168X_Fortune 文章写的非常深刻，最近高强度研究学习 @qinbafrank 老师的过往内容，把他们总结成了一个知识库，更有一个专属智能体，帮大家用Frank老师的框架看市场，限时免费试用中，欢迎大家体验反馈 - themarketbrew.com/kol/qinbafrank

8 28 92 129K 235

最近邮件与私信收到了好多国外友人的催稿，有生物老师，也有土耳其老哥的来信，希望我继续将项目做下去。 我也大受鼓舞，只是最近工作太忙，先做一个demo页面看看，但我保证会持续迭代这个项目 我希望AI技术真正能照进教育领域，给孩子们一些帮助，也希望有想法的伙伴贡献你的idea 或者欢迎各位老师提出你的需求 项目开源地址：github.com/cclank/cell-ar…

岚叔 @LufzzLiz

a month ago

复刻差不多了，给大家分享教程并开源代码，先看看我的录屏效果如何，主要看前两个，后面还没做😂 首先特别感谢下原作者@DilumSanjaya，创造出如此美妙的项目，让我大受震撼👍 【手把手教程】 1️⃣ 直接将原作者视频发给coex，让其复刻视频的网站。codex 会截图、抽帧、生图然后进行默默复刻 2️⃣

41 112 698 263K 1K

Shoutout to these 5 startup founders with $1M+ revenue and 100- followers. @VacarciucF @mathorde @TarkleHQ @SowaKillian @damjan_malis

19-YEAR-OLD GIRL HIT $43,000 IN SHOPIFY SALES IN 2 MONTHS. SHE RAN 1 STORE, 17 PRODUCTS TESTED, 6 WINNING ADS, AND HER ONLY “EMPLOYEE” WAS CLAUDE she wasn’t using claude to write product descriptions. that’s the beginner layer. she used it to study 12 competitors, pull patterns from 60+ bad reviews, build ad angles, write a 5-email flow, and check the store numbers every sunday. the first 3 weeks were ugly. around $4,200 spent on tests, 11 weak creatives killed, 1.8% conversion rate, and too many carts dying before checkout. then she pasted competitor reviews into claude and found the same complaint repeating across buyers: people didn’t trust the product would look the same when it arrived. that changed the whole store. new product page, clearer shipping copy, 4 trust blocks, better before/after visuals, and ad scripts built around the exact objection people already had. the winning creative did 3.6x ROAS while the old ones barely crossed 1.2x. by month two the system was cleaner. 43k in sales, 704 orders, 412 returning customers, five automated emails, weekly claude audits, and one simple rule: if an ad, product, or page section didn’t move numbers in 72 hours, it got cut. the store didn’t grow because “ai replaced shopify.” it grew because claude removed the guessing from every boring decision. read the full breakdown below if you still use claude like a description writer.

Gipp 🦅 @gippp69

a month ago

x.com/i/article/2056…

29 21 405 1.8M 1K