大王的大

别再用简单的脚本冒充 AI 交易机器人了。大多数所谓的 "AI Bot" 其实只是在大模型外面套了一层皮，根本无法处理复杂的交易逻辑。 真正的智能交易，需要的是有状态的推理和多智能体 协作。我用 LangGraph + Python 构建了一套纯开源的解决方案：LangTrader。它是为开发者设计的交易操作系统。👇 (1/5)

@boristane Time changed but many people refuse to accept due to significant cost loss of time and energy on software development they have spend in the past

people bragging about writing code manually are interesting

@VasiliyZukanov I guess maybe most AI labs and agencies frameworks are preferred python as the first language

Honest question: why choose Python for the backend in the age of AI? When agents write almost all the code, and human preferences become less important, wouldn't it make sense to optimize for performance, scalability and safety with languages like Java, C#, Go, etc.?

If you have experience that work with AI for long time, you can’t suffered any coding by hand making, it’s kind like used to trivial by plane but switch to riding a bicycle. Engineers who can address problems will thrive ,CURD boys or engineer don’t understand business logic gone

If your cursor or claude code bills never hit 200 pre month, don’t say you’re already understand vibe coding . If you’re the guy who believes 20/mo is enough, you’re out of the time bro, next generation programmers pattern is building, thinking out of the box.

@MaxMusing Our team is building AI driven saas or agents network, it’s very different design pattern , saas will remain for years but saas business by old pattern will decrease. First is coding next every aspect of white collar jobs will be handled by AI-SaaS, No one can stop it.

This has been said a thousand times before, but allow me to add my own voice: the era of humans writing code is over. Disturbing for those of us who identify as SWEs, but no less true. That's not to say SWEs don't have work to do, but writing syntax directly is not it.

@Web3Tinkle 厉害了，受nofx启发 开源一个类似的AI系统，底层基于langgraph+CCXT, 支持类似n8n那样的workflow配置，自己可以添加不同插件配置不同工作流，支持多Agents辩论投票，可以实现基于夏普率的自进化，欢迎体验：github.com/neilzhangpro/L…

1月5号我发布的，从第一性原理出发，研究ai交易的底层逻辑。 随后，基于底层原理，研发出来了识别regime的ai自适应网格交易，目前在内测中，相当的稳健😊。 图三，我们已经把行情模式分为36个子分类，每一个分类对应一种执行策略。 如果你能读懂我上一篇文章，就知道我们的方向是站在ai+量化双巨人的肩膀上前行，很少有人真正理解我们在做什么。

项目完全开源，无任何付费墙。 无论你是想寻找一个稳健的交易框架，还是想学习 LangChain/Agent 的高阶用法，这都是一个绝佳的参考案例。 🔗 GitHub: github.com/neilzhangpro/L… 欢迎 Star ⭐️，欢迎 PR。Let's build the future of algorithmic trading. #AI #开源 #量化交易 #Python #LangChain

这个项目非常适合作为 AI 辅助编程 (Vibe Coding) 的实战靶场。 你可以尝试让 AI： "写一个基于布林带突破的策略组件" "修改风控逻辑，当回撤超过 5% 时强制平仓" "接入一个新的新闻源 API" 架构已经搭好，你只需要填充你的创意。 (4/5)

别再用简单的脚本冒充 AI 交易机器人了。大多数所谓的 "AI Bot" 其实只是在大模型外面套了一层皮，根本无法处理复杂的交易逻辑。 真正的智能交易，需要的是有状态的推理和多智能体 协作。我用 LangGraph + Python 构建了一套纯开源的解决方案：LangTrader。它是为开发者设计的交易操作系统。👇 (1/5)

Mocking vibe coding today is just like mocking frontend devs as mere 'pixel pushers' a decade ago. Soon, they will witness the rise of an entirely new profession.

Initially, few noticed jobs were scarce & competition fierce. They claimed: "Just entry-level issue; mid/senior roles safe." Continued mocking AI hype. But as AI conquered once-secure intellectual peaks, no one escaped impact.

@adxtyahq I am a developer and I can’t agree with that. LLM will keep evolving and those issues will be addressed in the future. There are programmers demanding but advanced guys only. The world is changing bro

Vibe coding is fun until production shows up

@lifesinger 套到最后都是给AI喂料的

若干年后，AI 会发现：套壳的难度，远大于训练基础模型。套壳公司的市值，也将远超模型公司。 大部分人类不太信。这就是机会所在。

@quantgate Interesting! How do you see the bridge between high-level LLM reasoning and low-level microstructure signals? Would love to learn more about QuantGate's approach!

LangTrader sounds intriguing! At QuantGate, we're also pushing boundaries in quant trading with our autonomous system. Our focus is on real-time order-book signals for disciplined execution. Curious how this compares to LangGraph? Dive into microstructure with us! 📊🔍 #QuantTra...

AI Agent 炒币要来了？🤖📈 推荐 GitHub 上的 LangTrader。它不是简单的量化脚本，而是利用 LangGraph 构建的多智能体交易系统。 ✅ 核心：基于 LangChain 图结构实现复杂交易逻辑 ✅ 亮点：模拟“分析-决策-执行”的 Agent 协作链路 github.com/neilzhangpro/L…

@xDinoDeer 做生意角度看 死开发一直都是负债啊 能赚钱进来的才是资产

以 Claude code 发布为节点： 之前，技术团队是资产； 之后，技术团队是负债。

@akramcodez 💯

No worries, situation is under control