Promoting AI agents

Promoting AI agents

English

At the end of last year, AI agents really came alive for me. Partly because the models got better, but more so because we gave them the tools to take their capacity beyond pure reasoning. Now coding agents are controlling the terminal, running tests to validate their work, searching the web for documentation, and using web services with skills we taught them in plain English. Reality is fast catching the hype!

中文

去年年底，AI agent 对我来说真正活过来了。部分原因是模型变强了，但更重要的是我们给了它们工具，让它们的能力超越了纯粹的推理。现在编码 agent 能控制终端、运行测试来验证自己的工作、在网上搜索文档、并且使用我们用纯英语教给它们的技能来调用网络服务。现实正在快速追赶炒作！

English

This is all very evident if you've tried to employ any of the new models — especially Claude Opus 4.5, Codex 5, Gemini 3, and even the Chinese open-weight models like MiniMax M2.1 and GLM-4.7 — in one of the modern terminal harnesses that give them access to all these autonomous powers. The code being produced by this new breed of AI is leagues ahead of where their predecessors were at the beginning of 2025.

中文

如果你尝试过在现代终端工具中使用任何新模型——尤其是 Claude Opus 4.5、Codex 5、Gemini 3，甚至是中国的开源权重模型如 MiniMax M2.1 和 GLM-4.7——让它们获得所有这些自主能力，这一点会非常明显。这一代新 AI 产出的代码，已经远超它们前辈在 2025 年初的水平。

English

I've thoroughly enjoyed putting them all to work in OpenCode, which is a terminal interface for coding agents that allows you to seamlessly switch between all of the models, capture your sessions for sharing, and simply looks astounding when theme-matched with the rest of Omarchy (where we're making it a default in the next version!).

中文

我在 OpenCode 中充分体验了让这些模型协同工作的乐趣。OpenCode 是一个编码 agent 的终端界面，可以让你无缝切换所有模型、捕获会话用于分享，而且当它与 Omarchy 的其他部分主题匹配时，视觉效果令人惊叹（我们将在下一个版本中把它设为默认选项！）。

English

See, I never really cared much for the in-editor experience of having AI autocomplete your code as you were writing it. That was the original format pioneered by GitHub's Copilot and Cursor, but it left me cold. When I code, I want to finish my own thoughts and sentences. That was the sentiment I expressed on the Lex Fridman podcast last summer.

中文

你看，我从来不太喜欢在编辑器里让 AI 在你写代码时自动补全的那种体验。那是 GitHub Copilot 和 Cursor 开创的原始模式，但让我无感。当我写代码时，我想完成自己的想法和句子。这就是我去年夏天在 Lex Fridman 播客上表达的感受。

English

But with these autonomous agents, the experience is very different. It's more like working on a team and less like working with an overly-zealous pair programmer who can't stop stealing the keyboard to complete the code you were in the middle of writing. With a team of agents, they're doing their work autonomously, and I just review the final outcome, offer guidance when asked, and marvel at how this is possible at all.

中文

但有了这些自主 agent，体验完全不同了。它更像是在与一个团队合作，而不是和一个过度热情的结对程序员一起工作——后者总忍不住抢你的键盘来补全你正在写的代码。和一群 agent 合作时，它们自主完成工作，我只需要审查最终结果、在被要求时提供指导，然后惊叹这一切竟然真的可能。

English

Yes, I'm ready to give the current crop of AI agents a promotion. They're no longer just here to help me learn, answer my questions, or check my work. They're fully capable of producing production-grade contributions to real-life code bases.

中文

是的，我准备好给当前的 AI agent 们升职了。它们不再只是来帮我学习、回答问题或检查工作的。它们完全有能力为真实的生产代码库贡献生产级别的代码。

English

Yet pure vibe coding remains an aspirational dream for professional work for me, for now. Supervised collaboration, though, is here today. I've worked alongside agents to fix small bugs, finish substantial features, and get several drafts on major new initiatives. The paradigm shift finally feels real.

中文

然而，对我来说，纯粹的"氛围编程"（vibe coding）在专业工作中仍然是一个理想化的梦想，至少目前如此。但监督式协作已经到来了。我已经与 agent 一起修复小 bug、完成重要功能、并为几个重大新项目起草方案。范式转移终于感觉真实了。

English

Now, it all depends on what you're working on, and what your expectations are. The hype train keeps accelerating, and if you bought the pitch that we're five minutes away from putting all professional programmers out of a job, you'll be disappointed.

中文

当然，这一切取决于你在做什么、你的期望是什么。炒作列车不断加速，如果你相信了"我们离让所有专业程序员失业只有五分钟"的说辞，你会失望的。

English

I'm nowhere close to the claims of having agents write 90%+ of the code, as I see some boast about online. I don't know what code they're writing to hit those rates, but that's way off what I'm able to achieve, if I hold the line on quality and cohesion.

中文

我远没有达到网上有些人吹嘘的"agent 写了 90% 以上代码"的水平。我不知道他们写的是什么代码能达到这个比例，但如果我坚持质量和代码一致性的标准，那离我能达到的水平差得远。

English

But I'll forgive folks for getting excited! Because you don't have to connect many future dots on the current trend line to get dizzy by the prospects. The leaps of improvement that AI agents took in 2025 is simply incredible. This is the most exciting thing we've made computers do since we connected them to the internet back in the '90s. So what might things look like in 2026 or 2027? I get the exuberance.

中文

但我能理解大家的兴奋！因为沿着当前的趋势线，你不需要连接太多未来的点就会被前景搞得眼花缭乱。AI agent 在 2025 年取得的进步跨度简直不可思议。这是自 90 年代我们把电脑连上互联网以来，我们让计算机做的最令人兴奋的事情。那么 2026 或 2027 年会是什么样子呢？我理解这种狂热。

English

I also get that some programmers are eager to tune it all out. The hype drones on relentlessly, the most fantastical claims are still far off from being substantiated, and there's real uncertainty about where all this will leave the profession in the future. But that's still not reason enough to miss out on this incredible moment in human and computing history!

中文

我也理解一些程序员急于屏蔽这一切。炒作无休无止，最离谱的说法远未被证实，而这一切最终会让这个职业走向何方确实存在真正的不确定性。但这仍然不足以成为错过人类和计算历史上这个不可思议时刻的理由！

English

You gotta get in there. See where we're at now for yourself. Download OpenCode, throw some real work at Opus or the others, and relish the privilege of being alive during the days we taught the machines how to think.

中文

你得亲自下场。亲自看看我们现在到了什么程度。下载 OpenCode，把一些真实的工作交给 Opus 或其他模型，然后享受活在这个我们教会机器思考的时代这份特权吧。