集里怎么说它

• 《代码评审未死：人类从引擎变飞行员》(02:09起):本集说一旦合并前评审全是机器,观察代码上线后实际做了什么的生产环境就成了最后一个屹立不倒的审查者。

① 提到它的金句

48 条

但另一个现实是，你知道，构建一个 demo,在那种智能体世界或者说模型的世界里，是最简单的事情，但构建一个生产级的应用可能是最难的事情。
The other reality, though, is that, you know, building a demo, you know, in kind of the Agenda world or, you know, world of models is the easiest thing, but building a production grade app is probably the hardest thing.
—— Spiros · [09:39]

指向原始笔记的链接

如今生产环境中的智能体更像是自动驾驶汽车，我们也许有证据表明它比人类司机安全得多，但这仍然不足以让它完全放开，对吧？
Agency production today are more like self-driving cars, almost, where we might have proof that it’s a lot safer than a human driver, but still, this is not good enough to let it go, right?
—— Spiros · [24:03]

指向原始笔记的链接

我个人还没有见到很多用例，实际上在生产环境中还没有任何真正的智能体对智能体的用例，比如一群智能体的蜂群。
I have personally not seen many use cases, actually any use cases in production yet that are truly agent to agent, like a swarm of agents.
—— Jesse · [26:29]

指向原始笔记的链接

而对于我最关键任务的生产服务器，我仍然想要人在回路中。
And for my most mission critical production servers, I still want a human in the loop.
—— Vineet Edupuganti · [14:27]

指向原始笔记的链接

所以我觉得评估很重要，生产监控很重要，但是只有其中一个能为你解决问题的这种观点，在我看来是完全不屑一顾的。
So I feel evals are important, production monitoring is important, but this notion of only one of them is going to solve things for you that is completely dismissible in my opinion.
—— Aishwarya Reganti · [37:48]

指向原始笔记的链接

对其中很多平台来说,你可以非常快地搭建出一个东西,但如果你真的想把它投入生产或扩展规模,平台就到顶了。
With many of them you could build something very fast, but if you actually wanted to bring it into production or scale it, the platform maxed out.
—— Carlos González de Villaumbrosia · [06:35]

指向原始笔记的链接

地球上最复杂的组织通常最多也就是平均每月将软件交付到生产环境这种水平。
The most complex organizations on the planet usually deliver software to production on average monthly at best type of thing.
—— Chris Blackburn · [28:03]

指向原始笔记的链接

更难构建的能力是将那项前沿研究投入生产，并在安全关键的环境中部署它而不出现回归。
The much harder muscle to build is to carry that bleeding edge research into production and deploy it in a safety-critical environment without regressions.
—— Dmitri Dolgov · [22:56]

指向原始笔记的链接

你必须在生产内部使用 AI 来应对 AI 带入你的产品或系统的复杂性增加。
You’ve got to use AI inside of production to deal with the amount of increase of complexity that AI is putting into your product or into your system.
—— Justin Smith · [03:56]

指向原始笔记的链接

生产上下文重要得多，因为去检查仪表盘是一回事。说那个指标看起来不对劲是另一回事。
The production context is just way more important because it’s one thing to go check a dashboard. It’s another thing to say that metric smells off.
—— Justin Smith · [10:38]

指向原始笔记的链接

当你说,嘿,清除暂存数据库,而它在你的笔记本电脑上发现了一个它可以使用的令牌,并且它认为它正在与暂存一起工作,但实际上它是生产环境,现在它刚刚删除了一切。
when you say, hey, wipe the staging database and it finds a token on your laptop that it can use and it thinks it’s working with staging, but actually it’s production and now it just deleted everything.
—— Arjun Singh · [10:58]

指向原始笔记的链接

如果你没有一个好的基准测试，你就无法训练模型。而如果你不能训练模型，你就不需要在生产环境中部署它们。
If you don’t have a good benchmark, you can’t train models. And if you can’t train models, you don’t need to serve them in production.
—— Gabe Pereyra · [02:07]

指向原始笔记的链接

这就是拥有这种规模设施的优势——工程师们就坐在生产线旁边，而离生产线最近的团队，就是对生产线影响最大的团队。
Yeah, I mean, this is like the advantage of having a facility this size is that the engineers here are sitting right next to production and the teams that are sitting close to production are the ones who are most impactful to production.
—— Sorin · [05:24]

指向原始笔记的链接

所以显然在设计和生产过程中重度使用，但它确实也是我的技术联合创始人。
So obviously heavily in the design and production process, but also it’s literally my technical co-founder.
—— Yana Welinder · [03:14]

指向原始笔记的链接

每个人都在把智能体放进生产环境，如今甚至 15 岁的孩子都在这么做。该问的问题是它们是否真的有效。
Everyone is putting agents into production, even like 15-year-old kids these days. The question to ask is whether they actually work.
—— Ayush Bhardwaj · [03:17]

指向原始笔记的链接

每个 AI 都能到达生产环境，只是它没能跑起来，或者没能证明自身的成本合理。
Every AI reaches production, but it just fails to work or justify its own cost.
—— Ayush Bhardwaj · [16:37]

指向原始笔记的链接

而我见过走得好的地方是，如果你从一开始就认真对待一个可投产的、规模化企业系统的约束，把那些当作你要构建一切的架构原则，然后用你的新原语反向构建，重新达到那个 POC 的准确率。
But where I’ve seen it go well is if you take the constraints of a production-ready scaled enterprise system seriously from the beginning and treat those as the architectural principles that you’re going to build everything upon and then build back up towards that POC accuracy using your new primitives.
—— Saul Howard · [18:32]

指向原始笔记的链接

我们想在 2027 年达到一个地方：你拥有零宕机、生产环境中零 bug，至少是关键和高严重性的那些。
We want to reach in 2027 to a place where you have zero outages, zero bugs in production, at least the critical and the high ones.
—— Itamar Friedman · [18:24]

指向原始笔记的链接

一个真正的 bug 可以带着绿色勾选进入生产环境，而且直到很久以后都没人会注意到它。
A real bug can actually walk into production with green checks and nobody would notice it until much later.
—— Sachin Malhotra · [06:50]

指向原始笔记的链接

事实上，有一份 MIT 报告说，95% 的智能体项目无法从试点走向生产。
In fact, there was an MIT report that says 95% of agent projects are not able to move from pilots to production.
—— Manoj Saxena · [21:01]

指向原始笔记的链接

热和功、熵产生这些概念，我们谈论机器学习模型时并不使用，但我认为它们为思考正在发生什么、以及如何改进模型，提供了一个非常新颖且有趣的视角。
Concepts like heat and work and entropy, production, these things we don’t use when we talk about machine learning models, but I think they give you a very new and interesting perspective on how to think about what’s going on and how to also improve them.
—— Max Welling · [45:03]

指向原始笔记的链接

如果你有一套好的评估套件，你可以让 Claude、Codex 之类的进去替你修改提示词，而你确信你不会发布任何会毁掉生产环境的东西。
If you have a good eval suite, you can let Claude or Codex or Devon kind of go in, make prompt changes for you, let BlinkchainEngine make prompt changes for you, and you know that you’re not shipping anything that is going to ruin production.
—— Vishu · [02:26]

指向原始笔记的链接

在控制台上点来点去很适合搞清楚发生了什么,但这不是你把东西部署到生产环境的方式。
Clicking around on the console is great for being able to figure out what’s going on, but it’s not how you deploy things into production.
—— Mike Chambers · [06:30]

指向原始笔记的链接

很大比例——我想说可能高达约 50%——的 Shopify 的 pull request（每次改动生产系统时写的东西）现在不是由传统意义上做工程工作的工程师创建的，而是来自我们公司公共聊天里的对话。
It’s that a very large percentage, I want to say it’s probably up to about 50% of pull requests in Shopify, which again, pull requests every time you change the production system, you write a pull request, are created now not by engineers doing engineering work in the traditional sense, but out of conversations in our common company chat.
—— Tobi Lütke · [07:31]

指向原始笔记的链接

它准备好上生产了吗？而答案不可避免地是，我不知道。
Is it ready for production? And the answer is inevitably, I don’t know.
—— Jason · [03:00]

指向原始笔记的链接

你没法在 VLLM 或 SGLang 上运行这些基于扩散的 LLM。我们必须构建自己的服务引擎，而且我们能处理真实生产工作负载的很多复杂性。
You cannot run these diffusion-based LLMs on VLLM or SGLang. We had to build our own serving engine, and we can handle a lot of the complexity of real production workloads.
—— Stefano Ermon · [13:01]

指向原始笔记的链接

第一，前沿开放权重模型让生产级智能体在经济上可扩展。
First, Frontier Openweight models make production agents economically scalable.
—— Byung-Gon (Gon) Chun · [13:53]

指向原始笔记的链接

去年全球所有新增电力生产中有 86% 是可再生能源。
86% of all new electricity production worldwide last year was renewables.
—— Al Gore · [08:46]

指向原始笔记的链接

每一个投入生产的视频生成模型,在底层都用了一条复杂的提示词补全流水线。
every single video generation model in production uses a complex prompt completion pipeline under the hood.
—— Anastasis · [22:39]

指向原始笔记的链接

在合适的地方,我们不是用 LLM 替换检索和排序系统,而是把推理和理解蒸馏进某种生产级排序架构。
Where it seems right, we’re not replacing the retrieval and ranking system with an LLM, but rather distilling the reasoning and understanding into some production ranking architecture.
—— Raghav Saboo · [09:06]

指向原始笔记的链接

所以从根本上说，生产环境的每一次失误、也包括每一次亮眼表现——因为我认为朝正向爬坡也很有用——都会变成我们智能体框架的一个任务。
So fundamentally, every production miss or every production goodness as well, because I think it’s useful to hill climb in a positive direction, becomes a task for our agent framework.
—— Zubin Aysola · [12:21]

指向原始笔记的链接

任何真正的生产系统恐怕都不应该是一个黑暗工厂，至少在目前这个阶段是如此
any real production system should probably not be a dark factory, at least at this time
—— Liz Fong-Jones · [15:46]

指向原始笔记的链接

问题就来了，人们会凭感觉把他们的生产数据库搞垮，然后就直接被扔掉了。
The problem then comes is that people will vibe their production database into the ground and like it just gets thrown away.
—— David Burns · [41:12]

指向原始笔记的链接

我们仍然认为在进入生产环境之前需要有一个小小的人类在环中,至少目前是这样。我们都知道终极未来是要避免这一点。
We still believe there’s like some little human that needs to be in the loop before going to the production, at least at this point. We all know that the ultimate future is to avoid that.
—— Zilvinas · [10:25]

指向原始笔记的链接

人们对被推入生产的代码的理解却更少了。
People have less understanding of the code that’s being pushed into production.
—— Eric Schwartz · [02:11]

指向原始笔记的链接

在我加入 OpenAI 最早的时候，第三天我就把生产环境搞挂了。
Very, very earliest for me at OpenAI, I took production down day three.
—— Tibo Sottiaux · [30:25]

指向原始笔记的链接

就是这些硬技能——要把一个好点子变成产品所需的技能门槛太高了。
And it’s just like the hard skills, the creative skills required to get even a good idea into production are so high.
—— 嘉宾 · [28:03]

指向原始笔记的链接

我能告诉你的是，Armiden 自今年 2026 年 1 月以来，我们在客户站点发现了超过 90 个零日漏洞，全部在生产环境中。
And it’s like, all I can tell you is Armiden since January of this year, in 2026, we have found over 90 zero days at customer sites, all in production.
—— Kevin Mandia · [17:07]

指向原始笔记的链接

我认为二元评分，尤其是在起步阶段、当我们有一个正要投入生产的智能体时，比 A、B、C、D 之类要好。而且你应该极其严苛。
Binary scoring, I think, especially at the outset, when we have sort of an agent that we are putting out into production, binary scoring, I think, is better than, you know, A, B, C, D, or so on. And you should be incredibly harsh.
—— Doug Guthrie · [12:46]

指向原始笔记的链接

但现实中，在工程工作里，我们 70% 的时间花在另一头，不是生成新软件，而是在生产系统上，基本上就是运行和修复软件。
But in reality, for engineering, 70% of our time is spent on the other side, not generating new software, but rather in production systems, basically running and fixing software.
—— Varun Krovvidi · [02:31]

指向原始笔记的链接

我们真诚地相信，你在生产系统上所做的所有这些工作——比如修复问题或者日常运行你的软件——在大部分情况下都应该由智能体来完成。
So we genuinely believe that all of this work that you do on production systems, like fixing issues or running your software on a day-to-day basis, should be done by agents in most part.
—— Varun Krovvidi · [05:57]

指向原始笔记的链接

但一旦我把它发布到生产环境，我就是在盲目的状态下运行。
But once I ship it to production, I’m operating in the blind.
—— Hossein · [10:42]

指向原始笔记的链接

看，我觉得这时候我得搬出我的朋友 Charity，说真相是生产环境。其他一切都是我们讲给自己听的谎言。
See, I feel like I’ve got to channel my friend Charity at this point and say the truth is production. Everything else is the lies we tell ourselves.
—— Sam Newman · [35:12]

指向原始笔记的链接

这意味着，从我开始讲话的那一刻起，外面很可能就有一个 AI 攻击成功突破了一个生产系统。
It means that since the moment I began talking, there is probably an AI tech out there that was able to successfully bridge a production system.
—— Eli Cohen · [04:25]

指向原始笔记的链接

如果你拉取你的 Dora 指标，从代码写完到进入生产环境的时间，我敢打赌它远远超过实际做那些工程改动所花的时间。
If you pull your Dora metric, the time from code written to time in production, I would bet dwarfs the time actually making those engineering changes.
—— Davis Palmie · [05:38]

指向原始笔记的链接

每个月都有故事说某个智能体删了点东西，导致别人丢了生产数据库。
Every month there’s a story where someone lost their production database because the agent deleted something.
—— Andy Pavlo · [00:03]

指向原始笔记的链接

你不会把一个什么都做的微服务发布到生产环境，对吧？
You are not going to ship a microservice in production that does everything, right?
—— Viren Baraiya · [03:19]

指向原始笔记的链接

现在，技术的生产速度快于技术的消费速度。
Right now, the production of technology is faster than the consumption of technology.
—— Mike Cannon-Brookes · [31:32]

指向原始笔记的链接

② 出现在这些集

1 集

• 《代码评审未死：人类从引擎变飞行员》 — 作为概念

③ 关联

点进去有真内容 —— 本页主要出口

Laurie Voss · 代码审查 · 智能体 · 提示词注入 · 基准测试 · 误报 · GitHub · Cursor · OpenAI · Anthropic