Practical AI
Building the Foundation for the Agentic AI Era
In this episode of Practical AI, guest Angie Jones discusses her work at IBM, Twitter, and Block, where she led developer relations and helped create the Goose agent. She explains the formation of the Agentic AI Foundation under the Linux Foundation with OpenAI, Anthropic, and Block to provide neutral governance for protocols and standards such as MCP, AGENTS.md, Goose, and Agent Gateway. The conversation also covers how organizations can introduce agents to technical and non-technical teams, manage resistance, and identify where humans retain a distinct advantage.
本期 Practical AI 的嘉宾是 Angie Jones，她曾在 IBM、Twitter 和 Block 任职，负责 developer relations，并参与创建和推广 Goose。她介绍了 OpenAI、Anthropic 与 Block 在 Linux Foundation 下共同成立 Agentic AI Foundation，为 agentic AI 提供中立治理，并推动 MCP、AGENTS.md、Goose、Agent Gateway 等项目与标准。她认为，企业推广 AI agents 时不应只面向 developers，也要帮助 finance、marketing、HR 等非技术团队理解 AI 的强项与局限。她还强调，应让员工逐步把任务委派给 agents，同时保留人类具有明显优势的环节。面对快速变化和企业级复杂系统，需要通过种子用户试点、把有效做法嵌入 repo 和系统等方式扩大采用。
00:002712
00:01
Welcome to the Practical AI Podcast, where we break down the real-world applications of artificial intelligence
欢迎收听 Practical AI Podcast，我们在这里拆解 AI 在现实世界中的应用
00:08
and how it's shaping the way we live, work, and create.
以及它如何塑造我们生活、工作和创造的方式。
00:12
Our goal is to help make AI technology, practical, productive, and accessible to everyone.
我们的目标是让 AI 技术变得实用、高效，并且对所有人都触手可及。
00:18
Whether you're a developer, business leader, or just curious about the tech behind the buzz,
无论你是开发者、商业领袖，还是只是对这股热潮背后的技术感到好奇，
00:22
you're in the right place.
你来对地方了。
00:24
Be sure to connect with us on LinkedIn, X or Blue Sky
别忘了在 LinkedIn、X 或 Blue Sky 上关注我们，
00:27
to stay up to date with episode drops, behind the scenes, content, and AI insights.
以便及时获取新集上线、幕后内容以及 AI 见解。
00:32
You can learn more at practicalai.fm. Now, onto the show.
你可以在 practicalai.fm 了解更多。好了，节目开始。
00:41
Welcome to another edition of the Practical AI Podcast.
欢迎收听新一期的 Practical AI Podcast。
00:44
I am your co-host, going solo today. I'm Chris Benson. Daniel's not with me this time.
我是你的联合主持人，今天独自一人主持。我是 Chris Benson。这次 Daniel 没和我一起。
00:51
But we have an excellent conversation coming up for you.
但我们为你准备了一场非常精彩的对话。
00:54
What are you with me today? I have Angie Jones, who is the vice president of the Agentec AI Foundation,
今天和我一起的是谁呢？这位是 Angie Jones，她是 Agentec AI Foundation 的副总裁。
01:01
which I think is a super cool title to have at a super cool name place.
我觉得能在这个名字超酷的地方担任这个头衔，简直太酷了。
01:06
And I'm really looking forward to finding out more about it. Angie, welcome to the show.
我真的很期待了解更多。Angie，欢迎来到节目。
01:11
Thanks so much, Chris.
非常感谢，Chris。
01:12
So, like I said there in the intro, and I said it before the show started and stuff,
所以，就像我在开场白里说的，在节目开始之前我也说过这些什么的。
01:18
like, if somebody in AI was looking for a play like the Agentec,
比如说，如果有人在 AI 领域想找像 Agentec 那样的打法，
01:24
I mean, that is the coolest sounding thing you can have.
我是说，那听起来绝对是最酷的东西了。
01:29
But before we dive too far into the foundation, I'd really like to kind of hear,
但在我们深入基础之前，我真的很想听听，
01:35
like, how do you develop in your career so that you end up doing that?
比如，你是如何在职业生涯中发展到最终做这个的？
01:42
Like, could you tell us a little bit about your background?
比如，你能跟我们讲一下你的背景吗？
01:45
Because that's one of those things, if you wouldn't just said something to it,
因为这是那种，如果你不介意我说的话，
01:50
like, I work with tons of people working on Agentec AI,
比如，我和很多做 Agentec AI 的人一起工作，
01:53
but leaning the foundation of your goal, I'm just curious, how do you get to that point?
但在了解你的目标背景时，我特别好奇，你是怎么走到那一步的？
01:57
Yeah. So, I'm your traditional techie.
对，所以我就是个典型的技术人。
02:02
So, I've worked as an engineer for a couple of decades.
所以我当工程师有二十来年了。
02:06
So, I've placed this IBM and Twitter, and the last role was at Block
所以我待过 IBM 和 Twitter，最后一个职位是在 Block
02:13
in an engineering leadership capacity.
做的是工程领导这一块。
02:16
And so, in that role, one of my tasks was to basically teach the entire company,
所以在那份工作里，我的任务之一基本上就是教全公司
02:24
12,000 people how to use AI agents.
一万两千人怎么用 AI agents。
02:27
And this is as I'm learning myself, because I mean, there's no book for this,
而且这也是我自己在学，因为我是说，这东西没有参考书，
02:31
you know, right out the gate, and that was early like 2024.
你懂吧，从一开始就没有，那大概是2024年初。
02:36
So, a lot of this stuff was brand new.
所以，这些东西很多都是全新的。
02:38
It was not even common in tech, let alone in others' verticals, right?
这在科技领域都不常见，更别提其他垂直领域了，对吧？
02:43
And so, I also, like, lead developer relations.
所以，我也同时负责开发者关系。
02:47
And so, a big part of my role is helping developers worldwide understand new technologies
我的工作很大一部分是帮助世界各地的开发者理解新技术，
02:54
and how to use them.
以及如何使用它们。
02:56
And so, our company, Block, created like this internal AI agent,
所以，我们公司 Block 构建了一个内部 AI agent，
03:02
I named Goose, and Goose, we were the developers were using Goose
我给它取名 Goose，我们这些开发者当时就在用 Goose，
03:07
to like automate engineering tasks, help with coding and things like that.
来自动化工程任务、帮助写代码之类的。
03:12
And so, we were teaching developers across the globe, what is an AI agent?
所以，我们当时在教全球的开发者，什么是 AI agent？
03:16
Like, we were very early in this.
就像，我们做这个做得特别早。
03:18
And so, Jack Dorsey, who leads Block, was like, hey, Angie, you're teaching like everybody else about agents.
然后，Block 的负责人 Jack Dorsey 就说，嘿，Angie，你一直在教别人 agent 相关的东西。
03:27
And really would love for everyone in this company to learn how to use agents.
他真的很希望公司里每个人都能学会怎么用 agents。
03:32
And so, I'm talking finance, marketing, like, you know, design, HR, everyone,
所以，我说的就是财务、营销，你懂的，设计、HR，所有人，
03:39
needed to learn about AI in more specifically how to utilize agents.
都得学 AI，更具体地说是怎么利用 agents。
03:44
And so, I remember the team went really deep on this.
然后，我记得我们团队在这个上面搞得很深。
03:48
And then, it got to the point, like, pretty AI fluent company where everyone is comfortable using this.
后来，公司就变得相当精通 AI，大家用这些都非常顺手了。
03:56
I needed to go really deep on the engineering org.
我需要深入了解工程组织。
03:59
And so, that was my home.
所以，那就是我的主场。
04:02
Our engineers were using this, but we weren't seeing a big difference in like developer velocity, for example, right?
我们的工程师在用这个，但我们没有看到比如 developer velocity 方面有什么大的不同，对吧？
04:11
And so, we explored that and learned that, you know, we're only at the tip of the iceberg.
所以我们去探索了，然后发现，你知道，我们只看到了冰山一角。
04:17
We really could do a whole lot more to get to this autonomous engineering org, right?
我们真的可以做得更多，来达到这个 autonomous engineering org，对吧？
04:23
And so, I pretty much drank from the fire holes of like all of the news, all of the releases, everything that's going out.
我几乎是一股脑地吸收了所有的新闻、所有的发布、所有出来的东西。
04:31
And I consumed that kind of filter out a lot of the noise and then bring the things that are valuable to the engineers.
我吸收了这些，过滤掉很多噪音，再把有价值的东西带给工程师们。
04:39
And so, I would say like, I know a lot about this space.
所以，我想说，我对这个领域了解很多。
04:43
And also, at my time at Block, it worked on a protocol.
而且，我在 Block 的那段时间，它做了个 protocol。
04:48
This was like a cross-border money movement protocol.
这个 protocol 是跨境资金转移的。
04:51
So, those who don't know Block, that's the company, the finance tech company that houses square and cash-app, right?
如果有人不知道 Block，那是家金融科技公司，旗下有 square 和 cash-app，对吧？
04:58
So, money is our jam.
所以，钱就是我们的拿手好戏。
05:00
So, I worked on this protocol, never thought I would be a protocol girl.
所以，我当时就做这个 protocol，从没想过自己会是个 protocol 女孩。
05:04
But learned a lot about like just kind of open standards in how all of that works as well.
但也学到了很多关于开放标准以及这些东西是怎么运作的。
05:10
Also, really big and open source throughout my career.
而且，在我的整个职业生涯中，我一直非常看重 open source。
05:14
I always believed in open source, contribute to open source, like advocate for it, right?
我一直相信 open source、贡献 open source，并且倡导它，对吧？
05:20
So, all of that kind of came together in this perfect storm as open AI andthropic and Block wanted to form a foundation for agentic AI.
所以，这一切就像一场完美风暴一样汇聚在一起，因为 OpenAI、Anthropic 和 Block 想要为 agentic AI 成立一个基金会。
05:33
They understood that hey, some of these standards, some of these open source projects that we're coming up with,
他们明白，嘿，我们正在提出的这些标准、这些开源项目中的一部分，
05:40
we probably shouldn't be the sole authors or owners of these things, right?
我们可能不应该成为这些事情的唯一作者或拥有者，对吧？
05:46
MCP is a great example.
MCP 就是一个很好的例子。
05:48
So, MCP is the model context protocol.
所以，MCP 就是 model context protocol。
05:50
This is what agents use to connect to applications and tools, right?
这就是 agents 用来连接应用程序和工具的东西，对吧？
05:55
Which we saw across Block, everyone in Block needed MCP service to connect to whatever applications they were using.
我们在 Block 内部就看到了这一点，Block 的每个人都需要 MCP service 来连接他们正在使用的任何应用程序。
06:03
And so, like anthropic realized, yeah, this probably should live in a neutral home.
所以，就像 Anthropic 意识到的那样，是啊，这大概应该放在一个中立的地方。
06:09
And so, those three companies came together to form the agentic foundation under the Linux foundation.
然后呢，那三家公司就一起在 Linux foundation 下成立了 agentic foundation。
06:15
So, Linux foundation has been around for decades, everyone knows them.
Linux foundation 已经存在几十年了，大家都知道它。
06:19
And so, yeah, so this is a new foundation.
然后呢，对，所以这是一个新的 foundation。
06:22
So, once we stood this up, me and our head of open source came over to the foundation.
我们把它搭起来之后，我和我们的 open source 负责人就过来了这个 foundation。
06:30
And we thought it was so cool.
我们都觉得这太酷了。
06:32
We started working here full time.
我们开始在这里全职工作。
06:34
Very cool.
非常酷。
06:36
And I like the Jack Torsey name drop there.
还有，我喜欢你刚才提到 Jack Torsey 那一下。
06:39
That's pretty good.
挺不错的。
06:41
So, you actually, you were actually working directly with him, as well, along the way.
所以，你其实，你其实这一路都是直接和他一起工作的。
06:45
Yeah, that's right.
是的，没错。
06:47
So, I worked with Jack at Twitter.
所以，我在Twitter和Jack一起工作过。
06:50
And then he brought me over to Block when they started developing like these open source projects and protocol.
然后当Block开始开发这些open source项目和protocol的时候，他把我带了过去。
06:57
Very cool.
很酷。
06:58
I want to back up for a moment because I got a couple of questions from things that you brought up there.
我想稍微往回倒一下，因为你刚才提到的那些事，我有几个问题。
07:03
And one is the education, because I think educating the organizations that folks are in,
其中一个就是教育，因为我觉得，教育大家所在的组织，
07:10
like you were, you know, path finding along the same kind of task that a lot of organizations are trying to do right now.
就像你，你知道，正在做 path finding，正是很多组织现在都在尝试的那类任务。
07:18
And that is, you know, especially like every year recently,
而且，你知道，尤其是最近每年都这样，但特别是 2026 年，就进步的速度而言，你知道，简直快得离谱，还有 agentics 的兴起，你知道，去年就已经有了，但今年它彻底席卷了整个世界。
07:22
but especially 2026 has been just insanely, you know, fast in terms of the, you know, the level of progress
所以，现在每个组织都在应对这个。
07:29
and the onset of agentics, you know, they were there last year, but this year it just has taken over the world.
而你在“教育”这个理念上挑了大梁，不只是针对 developers，而是针对整个组织，还有组织里那些可能人们不太关注的部分，因为他们往往太专注于 developers 了。
07:37
And so, like every org is dealing with that now.
你能稍微谈谈创造这种改变是什么样的吗？关于——我打算把这个问题分开来问。
07:40
And you have taken point on this notion of education, not just for developers, but for, you know, the whole org
而且你在教育这件事上是主要牵头人，不只是对开发者，而是对，你知道，整个组织。
07:48
and kind of parts of the org that maybe people aren't thinking about as much because they tend to be very focused on developers.
还包括组织里那些大家可能不太会想到的部分，因为往往都太聚焦在开发者身上了。
07:55
Can you talk a little bit about what creating that kind of change looks like in terms of, and I'm going to separate it.
你能不能稍微谈谈，创造这种变化是什么样的，就是说——我打算把它分开来讲。
08:02
I want to ask about the non-developers first, like when you're going into finance and you're going into these other,
我想先问一下非开发者的情况，比如当你进入金融部门，进入这些其他部门时，
08:08
these other organizational departments that have, that are not thinking about the bits and bytes of AI all the time.
这些其他组织部门，他们并没有一直在思考 AI 的 bits and bytes。
08:15
And you're trying to say, here's a new tool, and like how do you approach that not only from the upskilling that's required,
然后你想说，这里有一个新工具，那么你如何应对这个问题，不仅是从所需的技能提升角度来看，
08:26
but also from the kind of the, like, getting people to accept it.
还有从那种，怎么说呢，让人们接受它的角度。
08:31
Because, I mean, you see people out there, there's a lot of resistance to AI in the general population out there.
因为，我的意思是，你看外面的人，普通大众对 AI 有很多抵触情绪。
08:37
And so, like, how do you navigate that when you're trying to move the org forward like that?
所以，当你想要推动组织前进时，你会如何应对这种情况呢？
08:44
Yeah, these were two very big and different challenges, right?
是的，这两个是非常大且不同的挑战，对吧？
08:50
So, one is the whole change management of it all, like you're essentially asking people to think differently and do their jobs differently.
所以，一个是整个 change management 的问题，就像你本质上是在要求人们以不同的方式思考和做事。
08:59
These are experts in their domains, right?
这些是各自领域的专家，对吧？
09:02
Who maybe been doing this a couple of decades and you're like, oh, kind of throw away like your processes and everything, and we want you to do this.
他们可能干这行已经几十年了，然后你突然说，哦，把你的流程啥的都扔了，我们要你这么做。
09:11
Not only that, to your point, resistance, right?
不仅如此，就像你说的，会有抵触，对吧？
09:15
Fear, lots of different emotions involved in that.
恐惧，还有各种复杂的情绪掺杂其中。
09:20
And then the whole technical part of it, where like these tools, like I said, this was very early on,
然后还有整个技术层面的东西，比如这些工具，像我说的，那会儿特别早期。
09:26
2024, 2025, you didn't have these nice desktop applications.
2024、2025年，还没有那些好用的桌面应用。
09:31
You had a CLI, right?
你只有一个CLI，对吧？
09:33
You were copying Jason to get an MCP server working.
你还得复制JSON才能让MCP server跑起来。
09:37
Like, this was foreign to folks who don't use these tools every day.
像是，这对那些不每天使用这些工具的人来说很陌生。
09:43
So, it was a really big job.
所以，那真是一项大工程。
09:47
And so, like, you have to consider all of that.
所以，你看，你得把所有那些都考虑进去。
09:50
It's not just, hey, let me show you how to use a terminal, right?
这不仅仅是，嘿，我教你怎么用 terminal，对吧？
09:54
It's helping them understand where they still fit into the process, right?
而是要帮助他们理解，他们在这个流程里还能起到什么作用，对吧？
09:59
With things, it's okay to let that go.
有些东西，该放手就放手。
10:02
Like, you probably should not be doing this anymore because it's not a good use of your time anymore, right?
像是，你可能不应该再做这个了，因为这已经不是合理利用你的时间了，对吧？
10:08
And then also, I like to say there's people on both spectrums.
而且我还想说，两种极端的人都有。
10:12
There's people that, you know, they think AI is the best thing ever and can do all the things.
你知道，有些人觉得AI是最棒的东西，什么都能做。
10:18
And there's other people who are like, AI is stupid, it can't do anything.
还有另一些人觉得AI很蠢，什么都不会。
10:22
I like to kind of be in the middle of there where, you know, I can be realistic about its strengths, its weaknesses,
我倾向于站在中间，你知道，能客观地看待它的长处和短处，
10:31
what I should use it for and what I shouldn't use it for.
知道该用它做什么，不该用它做什么。
10:34
And just leading with that helped a lot because I'm not trying to get you to drink Kool-Aid, right?
而且一开始就把这个立场亮出来很有帮助，因为我不是要让你喝Kool-Aid，对吧？
10:40
I'm saying, hey, here's a different approach that could speed you up, right?
我是说，嘿，这里有一种不一样的方法，能让你提速，对吧？
10:46
That can also give you access to a lot of stuff that maybe you didn't have before, and that car was the key.
而且它还能让你接触到很多以前可能接触不到的东西，而那辆车就是关键。
10:52
So when there's, let's say data, right?
所以当有，比如说，data，对吧？
10:56
Everybody needs data, no matter where you're in.
每个人都需要 data，不管你在哪里。
11:00
If I say, hey, listen, I can get you access with this agent and they're like some MCP server to let's say records that are in this database.
如果我说，嘿，听着，我可以让你通过这个 agent 访问，它们就像某种 MCP server，比如说这个 database 里的 records。
11:11
That you would have had to go to another team, put in a request and ask for a report, right?
你本来得去另一个团队，提交请求，然后要一份报告，对吧？
11:16
And then try to figure that report out yourself.
然后你还得自己琢磨那份报告。
11:19
I can get you that data in seconds so that you can go and like do your best work.
我几秒钟就能给你那个 data，让你可以去好好发挥。
11:25
So people are like sold on that kind of thing, right?
所以人们就会对这种东西很买账，对吧？
11:28
So you have to take them step by step and give them things that are meaningful to them that might not be their core job.
所以你得一步一步来，给他们一些对他们有意义、但不一定是他们核心工作的事情。
11:35
So just so they can get used to it and they could start building some trust, right?
这样他们就能慢慢适应，并且开始建立一些信任，对吧？
11:39
And then eventually they start delegating a little bit more and a little bit more until they find that right spot of, okay, I shouldn't delegate this.
然后最终他们开始一点一点地多交给AI一些，直到找到那个合适的点：好吧，这个不该交给AI。
11:48
The AI does not do well with that. That part I'll take, right?
AI不擅长那个。那部分我自己来，对吧？
11:53
So that was my overall approach to it.
所以这就是我的整体思路。
11:57
I really like that. And when I know when I get in conversations with folks about this topic, one of the things that I often say, which I think is kind of resonating with your story there,
我真的很喜欢这一点。而且当我和别人聊到这个话题的时候，我经常说的一句话，我觉得和你的故事挺呼应的，
12:08
is that like find the place where the human has a distinct advantage from the AI and differentiate the AI,
就是找到人类相对于AI有独特优势的地方，然后把AI区分开来，
12:16
you know, go down through the job requirements and find what, what are the human is still better at this point?
你知道，逐条过一遍工作要求，找出哪些是目前人类仍然更擅长的？
12:23
And that may be fluid. That may change over time depending on that.
而这个可能是动态的，可能会随着时间变化而变化。
12:26
So like having an open mind is good, but I think that's incredible advice that you're giving people in terms of like how to navigate.
所以保持开放的心态是好的，但我觉得你给人们的建议非常棒，关于如何应对这件事。
12:35
Because I think that's one of the biggest questions that people that people have out there based on the conversations I'm having.
因为我觉得，基于我平时参与的对话，这是大家最大的疑问之一。
12:42
To flip to the other side of the coin when you're dealing with developers and your and it has like even as if I'm a lifelong developer, I've been doing it for decades.
反过来说，当你在和开发者打交道时——就像我自己，我一辈子都是开发者，已经写了几十年了。
12:51
And I see, you know, and I, you know, every day on on social media and stuff, I see people.
而且你知道，我每天在社交媒体上，还有别的地方，都能看到人。
12:57
You see the developers that are embracing agents fully and they're the orchestra.
你会看到那些完全拥抱agents的开发者，他们就是那个指挥。
13:02
And then you see the ones that are like, oh, it's still sucks. And you know, I'm still, you know, all that stuff.
然后你也看到另一种人，他们会说，哦，这东西还是很烂。你知道，我还是那一套，就那些玩意儿。
13:08
And like, and how, what was your experience as you dived into the pool and developers trying to get the embrace going here?
那你跳进这个圈子里，和开发者一起试图让大家接受它的过程，你的体验是怎样的？
13:16
So for a while, if we look back, so back then you're like, oh, this is great.
所以有一段时间，回头想想，当时你会觉得，哦，这太棒了。
13:22
But it wasn't that great, right? I mean, and they're in the early days, it was good for that time.
但并没有那么好，对吧？我是说，早期的时候，以当时的标准是不错的。
13:29
But compared to now, it was not. And so developers who would like maybe try the tools and they didn't do a great job.
但跟现在比，那时候还不行。所以那些开发者，可能想试试这些工具，但效果不太好。
13:39
You ask it to write a feature or something. It's like, what? This is stupid.
你让它写个功能什么的，它就像是，什么鬼？这也太蠢了。
13:43
You know, those developers, I found once they tried it once if it didn't work out well, they kind of threw their hands up.
你知道，那些开发者，我发现他们只要试过一次，如果效果不好，基本就甩手不干了。
13:51
And it wasn't until like end of maybe 2025 when the like was it for like Sonic 45 or 46.
直到大概是2025年底，那时候好像是 Sonic 45 还是 46 之类的。
14:03
I don't even remember the versions anymore. But it was like this point in about November of 2025.
我都不记得版本号了。但大概就是2025年11月那个时间点。
14:09
The model is just really good.
那个模型就是真的很强。
14:12
But we couldn't wait until then. Like I had a mandate to like hit everybody using AI and to increase developer velocity.
但我们等不到那时候。比如我接到的任务就是让所有人都用上AI，提升 developer velocity。
14:20
And so what I did, thinking back to that change management.
所以我当时做的，回想起来，就是变革管理。
14:25
And it was too much of a lift to get 3500 developers to lift everyone at the same time.
要让 3500 个 developers 同时带动所有人，这个任务太重了。
14:34
Especially when you have all of these emotions involved and you have this resistance, right?
尤其是有这么多情绪因素，还有这种抵触情绪，对吧？
14:40
And I would say like we had quite a few people who were resistant to this.
我可以说，我们里有不少人对此很抵触。
14:45
But what I didn't stand, there's this this rule. It's called the the 190 rule.
但我不明白的是，有一条规则，叫做 190 rule。
14:52
And this rule essentially says that in any community, there's going to be 1% of that community that are creators.
这个规则本质上说，在任何社区里，都会有 1% 的人是创作者。
15:01
Think about like social media, anything like that. You'll have 9% that, you know, they'll dabble here and there, the tinkerers, if you will.
想想社交媒体之类的，会有 9% 的人，你知道，他们这里弄弄那里弄弄，算是爱鼓捣的人吧。
15:10
So the 90, the bulk of people are consumers, right?
所以那 90%，大多数人是消费者，对吧？
15:15
And so I said if I look at our engineering organization through that lands, let me go and put together the one print.
所以我说，如果用这个视角来看我们的 engineering organization，那我就去把那 1% 的人聚起来。
15:23
And so I went across the org and I pulled people and they didn't necessarily have to be the the drunk off the Kool-Aid like AI peeled people.
所以我走遍了整个组织去拉人，他们不一定要是那种像喝了Kool-Aid一样对AI狂热的人，但我只需要我们每个主要repo都有代表。你知道的，我们那些最大、最关键的项目，我想要来自不同团队、不同类型repo的人，所以我需要前端、后端、移动端，iOS和Android的人。你知道，我需要这些人。于是我就搞了个为期一周的小活动，走遍整个组织。这些人不一定——我不是他们的直属经理，对吧？所以我需要得到他们直属经理的支持。
15:34
But I just needed representation from every major repo that we had.
但我只是想让我们每一个主要的 repo 里都有代表。
15:41
You know, our largest ones, our critical ones, I wanted people from various teams, different types of repos as well.
你知道，我们最大的、最关键的 repo，我想要来自不同团队的人，不同类型的 repo 也要有。
15:50
So I needed front-end and back-end and mobile, iOS and Android.
所以我需要前端、后端、移动端，iOS 和 Android。
15:54
You know, I needed these people.
你知道，我需要这些人。
15:56
And so I did this little week-long campaign just going across the org.
于是我就搞了一个为期一周的小活动，在整个组织里跑了一遍。
16:00
These people don't necessarily, I'm not their first line manager, right?
这些人不一定，我不是他们的一线经理，对吧？
16:04
So I need to get buying from their first line manager.
所以我需要得到他们一线经理的支持。
16:06
That's a totally different conversation because a lot of them weren't even bought into this, right?
那完全是另一回事，因为很多人其实压根就不认同这个，对吧？
16:11
But I say, hey, listen, you know you have this mandate.
但我说，嘿，听着，你知道你有这个任务。
16:14
You've got to give people to use AI on your team.
你得让你团队里的人去用AI。
16:17
Give me one person that I can have 30% of their time.
给我一个人，让我能占用他30%的时间。
16:23
And what I want to do is, one, they all come together.
我想做的是，第一，让他们都聚在一起。
16:26
We learn all of this stuff, but it's not just for them.
我们学这些东西，但这不只是为了他们。
16:30
What I need them to do is build that knowledge back into the systems themselves so that your entire team benefits from this, right?
我需要他们做的是把这些知识反哺回系统本身，这样你整个团队都能受益，对吧？
16:41
And so they were there with me.
所以当时他们都跟我一起。
16:43
It was 50 of them, drank it from the fire holes, trying things out because things are changing so quickly.
他们50个人，从 firehose 里喝，尝试各种东西，因为变化太快了。
16:51
There's so much noise.
噪音太多了。
16:53
You don't know what'll work, what won't work, right?
你不知道什么能成，什么不能成，对吧？
16:56
Everything looks great on Twitter in a little demo.
在 Twitter 上的小 demo 里，一切都看起来很棒。
17:00
But we're working with huge, you know, monorepos with hundreds of services in them.
但我们处理的是巨大的，你懂的，monorepos，里面有几百个 services。
17:08
Like, you know, we're working with people's money.
比如，你知道，我们是在跟人们的钱打交道。
17:10
You know, these are enterprise systems that you have to approach this a bit differently and much more carefully.
你知道，这些是企业级系统，必须用不同的方式，而且要谨慎得多。
17:16
And so they would try things out, we'll see things, we'll try them.
所以他们就会尝试各种东西，我们看到了也会试。
17:21
Some things will work for maybe a specific type of repo.
有些东西可能只对特定类型的 repo 有效。
17:26
Like, okay, this works great on, well, sucks for mobile, right?
比如，好吧，这个在某些 repo 上很好用，但在 mobile 上就很烂，对吧？
17:30
We can't use that technique.
我们不能用那个技术。
17:32
But together, the 50 of us were able to like try a lot of things, figure out what works.
但我们 50 个人一起就能尝试很多东西，找出什么有效。
17:38
Maybe I do have something that works on my iOS repo.
也许我确实有一些东西能在我的 iOS repo 上起作用。
17:41
Hey, other iOS repels, here's what I figured out, right?
嘿，其他 iOS repo 的人，这就是我发现的，对吧？
17:45
And then, like I said, embed this stuff into the system.
然后，就像我说的，把这些东西嵌入到系统里。
17:48
So things like context engineering techniques, things like, you know, agents MD files and agent skills
所以像 context engineering techniques 这类东西，还有，你知道的，agents MD files 和 agent skills。
17:56
and all of that baked into the repo so that no matter who was pointing their agent app is repo,
所有这些都直接写进了 repo 里，这样不管谁把他们的 agent app 指向这个 repo，
18:02
the agent could work the way that your team wanted it to work because it was baked into the system.
agent 就能按你团队想要的方式工作，因为系统里已经内置好了。
18:08
So we didn't have to, you know, teach everybody how to do this.
所以我们不需要，你懂的，去教每一个人该怎么做。
18:11
These champions essentially did that for them and filtered out the noise for their team.
这些 champions 基本上替他们做了这件事，还帮团队过滤掉了噪音。
18:16
So they would bring back the things that actually work that are tried and true.
他们会把那些真正有效、经过验证的东西带回来。
18:20
And then they would do a brown bag session or something with their team.
然后他们会和团队搞一次 brown bag session 之类的。
18:24
Like, let me show you guys what I put in the repo and why it's working for you all of a sudden, you know?
比如，来，我给你们看看我往 repo 里放了什么，以及为什么它突然就对你们起作用了，明白吧？
18:30
So that was the technique that we used there and it worked really, really well.
这就是我们在那儿用的方法，效果真的非常非常好。
18:37
Sometimes as a business owner, it's so hard to put out quality material and content that looks handcrafted and human,
有时候作为企业主，要发布看起来像手工打造、充满人情味的高质量内容和素材，真的很难，
18:45
but yet still use appropriate tools and automation and AI to make that fast to ship.
但同时还是得用合适的工具、automation 和 AI 来快速上线。
18:52
That's why I appreciate so much what our partner Framer is doing.
这就是为什么我非常欣赏我们的合作伙伴 Framer 正在做的事情。
18:56
Framer is the pro website builder for creators, teams and businesses that want a professional site
Framer 是专业的 website builder，面向那些想要一个专业网站的创作者、团队和企业，
19:03
and care enough to get every detail right.
并且足够在意，把每个细节都做对。
19:06
But that can't come at the price of endless to-do lists and long, long development processes.
但这一切不能以没完没了的待办事项和漫长、漫长的开发流程为代价。
19:13
You need to make that happen fast, which is why Framer has integrated agents directly into their platform
你需要快速实现这一点，这就是为什么 Framer 将 agents 直接集成到他们的平台中，
19:20
that work right alongside the teams creating the site.
这些 agents 与正在创建网站的团队并肩协作。
19:24
You can learn more about Framer and learn how to get more out of your site
你可以了解更多关于 Framer 的资讯，学习如何更好地利用你的网站。
19:29
or get a Framer specialist or get started building for free today at framer.com slash practical AI
或者找一位 Framer 专家，或者今天就到 framer.com/practical AI 免费开始搭建。
19:36
for 30% off a Framer Pro annual plan.
即可享受 Framer Pro 年度计划 30% 的折扣。
19:39
That's framer.com slash practical AI for 30% off.
访问 framer.com/practical AI 即可享受 30% 折扣。
19:44
Framer.com slash practical AI rules and restrictions may apply.
framer.com/practical AI 可能适用相关规则与限制。
19:49
So that is super cool.
这真是太酷了。
19:51
I'm going to borrow your techniques myself.
我自己也要借用你的技巧了。
19:54
I'm going by a bit of a few other folks listening or watching will do the same.
我打赌，其他一些收听或收看的朋友也会这么做。
19:59
I'm really insightful there in terms of how to approach that.
关于如何处理那件事，我确实很有见地。
20:04
As you as you arrived having gone through a lot of this and you arrive at the
当你经历了这么多，然后来到 agente AI foundation，现在它是一个全新的组织。
20:09
agente AI foundation now and you're kind of it's brand new as an organization.
你正在把它从零搭建起来。
20:14
You're standing it up.
你在承担这些责任。
20:16
You're taking on the responsibilities.
你能聊聊把 foundation 运转起来、让它聚焦，找到合适的人，并知道自己在做那些需要做的事情——那些事情具体是什么，当时是什么感觉吗？
20:18
Could you talk a bit about like what was it like to get the foundation going
你能谈谈把这个基金会搞起来是什么感觉吗？
20:23
and get it focused and get the right people and know that you were
并让它有重点，找到合适的人，并且知道自己
20:28
doing the things that needed and what those are.
在做那些需要做的事情，以及那些是什么。
20:31
Yeah.
嗯。
20:32
So this is just a AI foundation was formed end of 2025.
所以这个AI基金会就是在2025年底成立的。
20:39
So it's been one about eight or nine months at this point.
所以到现在差不多八九个月了。
20:44
Lots of excitement, right?
很令人兴奋，对吧？
20:46
We had dozens of companies that were interested in being apart
有几十家公司都有兴趣参与进来，
20:51
because everyone sees the need for this.
因为大家都看到了这个需求。
20:54
Everyone is also innovating at lightning speed.
而且大家都在以闪电般的速度创新。
20:58
There's so much innovation that's happening right now.
现在正在发生的创新太多了。
21:01
And if you have anything that you're creating that's like, you know,
如果你正在创造的东西是那种，你知道的，
21:05
we don't have this figured out.
我们还没搞清楚这个。
21:07
We need a new standard for this.
我们需要一个新的标准。
21:09
No one wants that standard being cooked in one kitchen, right?
没人想要那个标准是在一个厨房里做出来的，对吧？
21:15
If we're talking about agente commerce, for example,
比如说如果我们讨论agente commerce，
21:18
PayPal doesn't want strike figure net out by themselves, right?
PayPal并不想让Strike自己搞定，对吧？
21:21
They want to work together or they need to work together.
他们想要合作，或者说他们需要合作。
21:24
And they know they need to work together.
而且他们知道他们需要合作。
21:26
And so what the agente AI foundation has done is provided this neutral home
所以 agente AI foundation 所做的就是提供了一个中立的家
21:31
where all of these various companies can come and have these conversations together.
在这里，所有这些不同的公司都可以聚在一起进行这些对话。
21:36
So we have working groups essentially where the members,
所以我们基本上有工作组，成员们，
21:41
they, okay, which working group do you want to be a part of?
他们会说，好吧，你想加入哪个工作组？
21:44
And there's so many of them.
而且有很多很多个。
21:45
There's like, like I said, the agente commerce, right?
就有像，就像我说的，agente commerce，对吧？
21:47
So if you're figuring out how agents are going to buy stuff on the web,
所以如果你是想弄清楚 agent 将如何在网上买东西，
21:50
all right, I need all I need bees.
好吧，我需要所有我需要的蜜蜂。
21:53
I need PayPal.
我需要 PayPal。
21:55
I need Stripe.
我需要 Stripe。
21:56
I need, you know, all of these various companies together.
我需要，你知道，所有这些不同的公司一起。
21:59
And they work on these standards together.
它们一起合作制定这些标准。
22:01
I like to think of it as they are defining the rules of the game
我喜欢这样想：它们正在共同定义游戏规则，并一起搭建游戏棋盘。
22:05
and creating the game board together.
一旦它们把这事儿搞定了，就像说，好了，现在我们可以竞争了。
22:08
And once they have that figured out, it's like, all right, now we can compete.
你知道，算我一个，对吧？
22:12
You know, deal me in, right?
你知道，算我一个，对吧？
22:14
And so that's what the agente AI foundation does.
所以，这就是agente AI foundation在做的事情。
22:18
It also houses those standards or those projects.
它也承载了那些标准或那些项目。
22:21
So right now, the projects are MCP, agents MD, goose, agent gateway.
所以现在，这些项目就是MCP、agents MD、goose、agent gateway。
22:29
And then also just this week, A2A, which is Google's agent to agent protocol,
然后就在这周，A2A——也就是Google的agent to agent协议——也加入了foundation。
22:36
has come into the foundation.
然后这些项目，现在有来自世界各地的公司在为这些项目做贡献，对吧，或者为这些标准做贡献。
22:38
And so these projects, now you have companies from all over contributing
然后你就给它，就像，你知道的，更多的话语权。
22:46
to these projects, right, or these standards.
参与这些项目，或者这些标准，对吧？
22:49
You then give it like, you know, more of a voice.
然后你给它，你知道，更多的话语权。
22:53
If you, if you wanted to use one of these projects, but it lives in one of these companies,
如果你，如果你想用这些项目之一，但它属于这些公司之一，
22:59
you might be a bit hesitant, right?
你可能会有点犹豫，对吧？
23:01
As a company to build your products on top of this.
作为公司，要在它上面构建你的产品。
23:05
If it lives with this one company, I don't know what they're going to do.
如果它属于这一家公司，我不知道他们会怎么做。
23:08
I don't know if they'll like kill it.
我不知道他们会不会直接把它砍掉。
23:10
I don't know if they'll just, the roadmap will only reflect their goals, you know what I mean?
我不知道他们会不会，路线图只会反映他们自己的目标，你懂我意思吗？
23:17
And so this being a part of the agente AI foundation gives it that confidence that, hey,
所以，作为 agente AI foundation 的一部分，这给了你信心，嘿，
23:23
this is in a neutral place, I now can build on top of this.
它在一个中立的地方，我现在可以基于它来构建了。
23:27
And so that's what it looks like.
所以，就是这个样子。
23:29
We do a lot of education around the projects, the protocols,
我们会做很多关于这些项目和 protocols 的教育和推广工作，
23:33
we throw conferences in all parts of the globe.
我们会在全球各地举办大会。
23:37
And I'll say like one, one really cool aspect of this role.
而且，我觉得这个角色有一个特别酷的地方。
23:42
So in my last role, when deep on agents, but I'm also inside of my company, right?
所以在我上一个职位，我深入做 agents，但也是在自己公司内部，对吧？
23:48
A little bit of open source.
也涉及一点 open source。
23:50
I'm consuming, but I'm still, my bubble is probably North America.
我虽然在使用，但我的圈子大概还是北美。
23:55
Now this is a global foundation where my job is to pay attention to what is happening
而现在这是一个全球性基金会，我的工作就是关注正在发生的事情。
24:02
across the world.
整个世界。
24:03
How are they adopting agente AI in Japan, in China, in Africa?
他们在 Japan、China、Africa 是怎么采用 agente AI 的？
24:10
You know what I mean?
你懂我意思吗？
24:11
And so now I have this global perspective, which is absolutely fascinating.
所以现在我有这种全球视角，真的超级有意思。
24:17
And part of the role is also to help these various countries come together on these standards
而且这个角色的一部分也是帮助这些不同的国家在 standards 上达成一致
24:25
and protocols and projects so that they're interoperable across the board.
以及 protocols 和项目，以便它们在整体上 interoperable。
24:30
And I'm curious to that, that does sound super cool.
我对此很好奇，那听起来超级酷。
24:33
I'm curious with that global perspective that you have developed in this role,
我很好奇的是，你在这个角色中形成的全球视角，
24:39
like how do you assess?
比如說你怎麼評估？
24:43
I think it's very easy for pretty much, I think it's easy for anyone kind of coming
我認為這對幾乎任何人來說都很容易，我是說，任何人
24:49
from their own perspective and their own little bubble they're in to kind of assume I'm doing
從自己的視角和自己所在的小圈圈出發，去假設我正在
24:55
agentics and everybody else is doing it just like me or has the same needs and stuff like that.
做 agentics，而其他所有人也都像我一樣在做這件事，或者有一樣的需求，諸如此類。
25:00
And so I think it's very easy to forget that diversity of location,
所以我認為很容易忘記地理位置的多樣性，
25:06
diversity of life, all those things can change the user's need for that.
生活型態的多樣性，這些都會改變用戶對它的需求。
25:11
And so do you have any any particular highlights from that global perspective on things?
那你從全球視角來看，有沒有什麼特別的亮點？
25:17
Maybe this surprised you or that caught your interest?
也許是讓你驚訝的事，或者引起你興趣的事？
25:21
I'd love to hear some of that.
我很想听听这部分。
25:22
So I would say in places like China that are mobile first, right?
所以我觉得，在像中国这种移动优先的地方，对吧？
25:29
The way they use technology is a bit different than us.
他们使用技术的方式和我们有点不一样。
25:34
So we're in North America, I would say a very like SASS heavy culture,
而我们在北美，我觉得是一种非常重 SaaS 的文化，
25:41
whereas everything is like mobile over there.
而那边一切都是以移动端为主。
25:45
And so now like they, for example, they have a big need for a protocol like agent to agent
所以现在，比如说，他们对于 agent 到 agent 这样的协议有很大需求。
25:52
where you need like this app to be able to talk to that app and they're both agentic apps, right?
就是需要这个 app 能和那个 app 对话，而且它们都是 agentic apps，对吧？
25:58
And so they could like use that like, for example, like we chat, uses that, you know,
所以他们就可能用得上这个，比如像 WeChat 就会用到，你懂的。
26:04
with various agents and stuff like that.
用了各种 agents 之类的。
26:06
So that was really fascinating to me to just see like,
所以对我来说，光是看到这个就真的觉得特别有意思，比如说，
26:10
even the types of applications you're using essentially influence like the types of tech that are standards
甚至你用的 applications 类型基本上会影响哪些 tech 会成为 standards 之类的，
26:20
or anything like that that you might need.
或者你可能会需要的任何类似的东西。
26:22
And so that's just one example.
所以这只是一个例子。
26:24
Yeah.
对。
26:25
That's cool.
真酷。
26:26
I'm curious.
我很好奇。
26:27
And this is where I'm going to be selfish on my part.
这就是我要自私一点的地方。
26:29
I'm as an engineer and a research scientist.
我既是工程师也是研究科学家。
26:33
I'm very focused on autonomy.
我非常专注于autonomy。
26:35
And as you're getting into mobile and that's getting awfully close to thinking about edge concerns and stuff like that.
而且当你进入mobile领域，就非常接近要考虑edge问题了，等等。
26:42
And in embodied intelligence, you know, is such a hot area.
而embodied intelligence，你知道，是一个非常热门的领域。
26:47
It's certainly the area that I'm focused on.
这当然是我专注的领域。
26:50
And you know, the rapid rise of robotics in all domains, you know,
而且你知道，robotics在各个领域的快速崛起，你知道的，
26:56
whether they're ground robots or flying things or whatever is truly taking off.
无论是ground robots还是会飞的东西，或者别的什么，真的在腾飞。
27:02
Like it never has before, no pun intended there.
就像以前从来没有过的那样，这里没有双关的意思。
27:06
I'm curious how with this kind of, and I think in some parts of the world you see that more than others.
我很好奇，在这种……而且我觉得在世界上有些地方，你看到的情况会比别的地方多。
27:13
I think in China, Japan, and there are certain places where you're going to see a lot more robotics than you will in the US.
我觉得在中国、日本，还有某些地方，你会看到比在美国多得多的 robotics。
27:20
For US listeners or watchers, we don't have nearly as much of that here.
对于美国的听众或观众来说，我们这里远远没有那么多的这种东西。
27:24
That's right.
没错。
27:25
And so, like, how does, when you're looking from trying to take these projects that the foundation has,
那么，比如说，当你在尝试接手基金会现有的这些项目时，
27:34
and you're looking at those different needs and you're seeing, well, you know, the Americans and the Canadians and such, you know,
然后你看看那些不同的需求，你就会发现，嗯，你知道，美国人、加拿大人那些人，你懂的，
27:40
kind of have one way of doing it.
有点像是同一种做法。
27:42
And the Chinese and the Japanese and other, you know, like that may have a different way.
而且像中国、日本这些地方，你知道，做法可能不太一样。
27:46
How do you reckon, because that's quite different, you know, in terms of how you're using agents in those ways.
那你觉得呢？因为你知道，在使用agents的方式上，差别还挺大的。
27:52
And the way you're configuring it and the way you're constructing them and what their utility is.
以及你怎么配置它们、怎么构建它们，还有它们的utility是什么。
27:57
How do you approach that when you're trying to make everybody happy with standards that are truly meant to be global?
那当你试图让每个人都满意这些真正面向全球的标准时，你是怎么处理的呢？
28:04
Yeah, that is why you need everyone at the table, right?
对，这就是为什么你需要让所有人都参与进来，对吧？
28:08
And so, if you only had, like, your top, like, fame companies in the US, kind of determining all of this.
所以，如果只有美国那些顶尖的、有名的公司来主导这一切的话。
28:17
And there's no one from these other countries that, hey, robotics is like a big deal here.
而且没有来自这些国家的人说，嘿，robotics在这里可是大事。
28:24
Like, we have to think about the physical applications as well, right?
比如说，我们也得考虑physical applications，对吧？
28:29
Then you miss that perspective, right?
那你就会错过那个视角，对吧？
28:31
And not to say, like, they don't have places there, but you need leadership from those companies to also have a part in crafting the story, right?
也不是说他们在那边没有位置，但你需要那些公司的领导层也参与进来，共同塑造这个叙事，对吧？
28:41
And so, that's exactly, like, what the working groups do, like, find your lane, get into it, you know, Europe is another good example.
而这正是工作组在做的事情，找到你的赛道，然后深入进去，你看，欧洲就是另一个好例子。
28:51
So, they just rolled out the EU AI Act, right? And so, that, like, is going to change the transparency that these systems has had.
他们刚刚推出了EU AI Act，对吧？所以这将会改变这些系统以往的透明度。
29:01
And that's global. If you're going to be serving anything to someone in Europe, then you have to be able to adhere to these new legislations.
这是全球性的。如果你要向欧洲的用户提供服务，就必须遵守这些新的法规。
29:10
And so, that's another example where, okay, I don't care what country you come from.
所以，这又是一个例子，说明，我才不管你来自哪个国家。
29:15
You have to, one, be aware of this, and two, also help define how this should be done, right?
你必须，第一，意识到这一点；第二，还要帮助定义这件事应该怎么做，对吧？
29:23
There was a lot of backlash on how anthropic growth is out, for example, right?
比如，关于Anthropic增长被曝光这件事，有很多反对声音，对吧？
29:29
And so, we have a working group that is looking at, are there some systems that we could build or some standards or something?
所以，我们有一个工作组在研究，能不能建立一些系统或者制定一些标准之类的？
29:39
So, that everyone doesn't have to reinvent this and figure out how they're going to do it, right?
这样大家就不必都从头开始，还得自己想怎么搞，对吧？
29:44
So, then you have the voices from various countries together to figure this out and say, okay, what if we did it this way?
然后，来自不同国家的人聚在一起想办法，说，好吧，如果我们这么做会怎么样？
29:51
They haven't solved that yet, by the way, but that's an, you know, an active group that's meeting regularly.
顺便说一句，他们还没解决这个问题，但你知道，那是一个很活跃的组，定期开会。
29:57
And by the way, all of these working groups are open, so anyone can join them.
而且，所有这些工作组都是开放的，任何人都可以加入。
30:03
Just, you know, they have meetings, they're public, you could just, like, kind of go into it, see what they're talking about, even, you know, chiming with thoughts of your own.
就是，你知道，他们有公开的会议，你可以直接进去，看看他们在聊什么，甚至也可以发表一下自己的想法。
30:13
But these folks meet, and this is, you have one for security, you have one for identity, like, you know, any lane that you care about,
但这些人确实在开会，而且，有负责安全的，有负责身份的，就是，你知道，任何你关心的领域，
30:22
there's someone there, like, kind of working on this from across the globe.
都有人在，比如，来自全球各地的人在做这方面的工作。
30:27
Yeah, I know, as a, very specific to, to that, it was just a few days ago, then thethropic released their paper, along with the blog post on watermarking,
是啊，我知道，具体来说，就在几天前，Anthropic 发布了他们的论文，还有那篇关于 watermarking 的博客文章。
30:37
so that you can detect, you know, AI generated content and all that, which was a fact, and like, if, if, where, this is probably not the only episode that's going to come up, why not?
这样你就能检测，你懂的，AI 生成的内容之类的。这确实是事实。而且，这个话题可能不会只在这期节目里出现，为什么不聊呢？
30:46
I think we're, we're going to talk, we're going to probably do a deep dive on that very soon, and hint.
我觉得我们很快会深入聊这个，而且，提示一下。
30:52
So, but that, that is one of those things that, you know, to your point, like, affects everybody, and, and they're probably needs to be somewhat universal approaches to what, what is watermarking?
所以，这种事情，你懂的，就像你说的，影响所有人，而且可能需要有某种通用方法来定义 watermarking 是什么。
31:06
How should it be applied?
应该怎么应用？
31:07
What should it be applied to? How do you, you know, how do you utilize that, both as the provider and as the consumer of that?
应该用在什么上面？作为提供方和消费方，你该如何利用它？
31:15
Or do you even do watermarking? Or do you do a better way to do that, you know?
还是说根本不做 watermarking？或者你有更好的方式？你懂的。
31:19
That's right. That's right. So, I'm, I think I, I'm starting to see lots and lots of different utilities.
没错，没错。我觉得我开始看到很多很多不同的应用场景了。
31:27
I, I imagine the foundations can grow quite a lot in the, in the years to come.
我，我觉得在未来几年里，foundation models 还能成长很多。
31:32
Yeah. That's definitely a lot of work to do.
是的，那确实有很多工作要做。
31:36
Super interesting and exciting to think about this agentic future driven by things like MCP and agent to agent interactions, but actually putting that architecture in place within your company within your enterprise can be overwhelming because it sometimes seems like you just lose complete control.
想到由 MCP 和 agent 与 agent 之间的交互所驱动的这个 agentic future，真的超级有趣、让人兴奋。但要在你公司内部、企业里面真正把这个 architecture 落实下来，可能会让人很头大，因为有时候感觉就像完全失去了控制。
31:58
That's why I'm so privileged to be leading a company prediction guard that is helping you gain that control back.
这就是为什么我非常荣幸能领导 prediction guard 这家公司，它正在帮你重新夺回那种掌控感。
32:05
By deploying prediction guard, which is a self-hosted AI control plane within your organization, you can manage the supply chain on which your agents operate, institute runtime governance, manage observability, and none of that slows down agent building because it ships with a robust agent builder called agent forge out of the box.
通过部署 prediction guard——一个在你组织内部的 self-hosted AI control plane——你就能管理你的 agents 运行所依赖的 supply chain，建立 runtime governance，管理 observability。而且这些都不会拖慢 agent building 的速度，因为它开箱就带一个强大的 agent builder，叫 agent forge。
32:27
I would encourage you to check out what we're doing at prediction guard dot com slash practical AI book a time to get a demo and talk with our team prediction guard dot com slash practical AI.
我建议你去 prediction guard dot com slash practical AI 看看我们在做什么，预约个时间获取 demo，和我们的团队聊聊。prediction guard dot com slash practical AI。
32:40
So as we, as we start getting to the point where we can, I'd love to dive into kind of maybe get a high level overview of those of each of those projects that you described a little while ago, just kind of what it is for people that are familiar with them.
所以当我们，当我们开始可以深入的时候，我想先大概了解一下你刚才介绍的那些项目，就是给熟悉它们的人讲讲它们是什么。
32:55
And then and then maybe dive into like where some of the stuff is going, you know, based on where you're at, I know that you and I have talked about MCP and other things.
然后再，再深入聊聊这些东西接下来会往哪里去，你知道，基于你目前的进展。我知道你和我之前聊过 MCP 和其他事情。
33:05
So I loved it. I'd love to go wherever you want to go in terms of some technical deep dives because we love back here.
所以我特别喜欢。你们想往哪走都行，聊多深的 technical deep dives 都可以，因为我们在后台都很喜欢。
33:12
Yeah, sure. So we have agents of D. So this was a standard created by open AI. And this one standardizes like how projects, meaning like co basis communicate their their operating instructions right to agents and then there's
嗯，当然。我们有 AGENTS.md。这是 OpenAI 创建的一个标准。它规范了项目，也就是 codebases，如何把 operating instructions 直接传达给 agents。然后还有
33:33
Goose, which is one of the first open source AI agents and this one essentially provides like this agentic runtime. And then there's MCP of course, which connects agents to tools and systems.
Goose，最早的 open source AI agents 之一，它基本上提供了 agentic runtime。然后当然还有 MCP，它把 agents 连接到 tools 和 systems。
33:51
Agent Gateway, this one was, oh, and I didn't tell you who's came from block and MCP came from in thropic agent gateway came from a company called solo and solo created agent gateway to mediate traffic.
Agent Gateway，这个——哦，我还没说 Goose 来自 Block，MCP 来自 Anthropic。Agent Gateway 来自一家叫 Solo 的公司，Solo 创建 Agent Gateway 是为了管理 traffic。
34:09
Essentially like MCP traffic a two a as you're using this stuff and enterprise, you kind of want some controls around what the agent is doing and what it has access to. You want to, you know, make that observable and stuff like that. So agent gateway is open source project that does that.
本质上，像 MCP traffic、A2A，当你在 enterprise 环境里用这些东西时，你会希望对 agent 在做什么、能访问什么有一些控制。你想让它 observable，诸如此类。所以 Agent Gateway 就是做这个的 open source project。
34:28
And then a two a is the newest addition of this one is from Google a two a is the agent agent protocol and this one is really cool and that this is how agents can coordinate with other agents. So you can delegate work to another agent or, you know, just be able to collaborate on a given task, right.
然后 A2A 是最新的一个，来自 Google。A2A 就是 agent-agent protocol，这个真的很酷，它让 agents 可以和其他 agents 协调。所以你可以把工作委托给另一个 agent，或者就是能针对某个给定任务协作，对吧。
34:51
That's pretty cool. The I'm curious, you know, one of the things that that I know in my world that I've been exploring is, you know, if you look at late last year and people would just have an agent, you know, to do something.
那挺酷的。我很好奇——你知道，在我这个圈子里我一直在探索的是——你看去年年底，大家还只是有一个 agent 去做某件事。
35:04
And then you know, we kind of moved into this year and people are doing multiple agents that are starting to collaborate a bit and you're starting to have that kind of crosstalk in collaboration.
然后呢，到了今年，大家开始用多个 agents，它们开始有点协作，你开始看到协作中那种 crosstalk。
35:14
And then now we, you know, as we record this, we're in August and and we're talking like it's gone orders of magnitude in terms of it's it's some and like that's not everybody.
然后现在，你知道，我们录这段的时候是八月，我们说起来这个已经提升了几个数量级，但这不是所有人都这样。
35:26
This is some use cases where you're talking tens of thousands of agents sometimes or hundreds of thousands or even millions are now coming into play.
在某些 use cases 里，你谈论的是几万个 agents，有时是几十万甚至几百万个 agents 现在开始发挥作用。
35:34
And as you're I'm curious like as you're looking at that has to be a challenge to some degree from a scalability standpoint because, you know, going back to what you were saying before we're we're racing along so fast, much faster.
而且，当你看着这个的时候，我很好奇，从 scalability 的角度来看，这在一定程度上肯定是个挑战，因为，你知道，回到你之前说的，我们跑得太快了，快得多。
35:51
And I'm old enough to remember like I won't I'm old and therefore I remember before internet was even a thing.
而且我年纪够大了，记得——我不会……我老了，所以我记得在互联网还没出现之前的日子。
36:00
And I was even an adult at that point sadly so.
而且可悲的是，那时候我甚至已经是成年人了。
36:05
And so like I know the velocity that things happened across each of the various technological revolutions that we've had over the past lifetime, if you will.
所以，我了解过去这一生中经历的各种技术革命中事情发生的速度，如果你愿意这么说的话。
36:15
And so this is going so much faster than any of those others.
而这次的速度比以往任何一次都要快得多。
36:20
And all of those have had processes, you know, that's why the lyrics foundation itself, you know, came into being in in a lot of others where these technologies came there was the need to get collaboration across competitors and across global concerns and needs.
而所有这些都经历过这样的过程，你知道，这就是为什么 Linux Foundation 本身会诞生。
36:38
But you've stepped into the fastest moving area ever and it's gone in just a space of months from like singles to millions.
但你现在踏入了有史以来变化最快的领域，短短几个月，就从个位数发展到了数百万。
36:49
And as we like, how does how does a working group and especially considering that these are at the end of the day all competitors, these different members, you know, in their in their own businesses.
而且我们就像，一个工作组怎么，尤其是考虑到这些人说到底都是竞争对手，这些不同的成员，你知道，各有各的业务。
37:00
How do you manage that because by the time working groups like historically by the time a working group would arrive at something like sometimes you were so far past that.
你怎么管理这个？因为到工作组的时候，历史上来说，等工作组最终达成一个共识的时候，有时候早就已经过时了。
37:09
So how how does is there anything, you know, like are there any expected time scales or anything on how to get these things together so that it stays relevant because the speed of relevance is just unimaginable now.
所以有没有，你知道，有没有什么预期的时间表之类的，怎么把这些东西凑到一起，让它保持相关性？因为现在相关性的速度简直不可想象。
37:24
So in terms of how fast that is, like how does this new foundation address these kinds of of problems that we've we've had to a lesser extent but never like today.
那么从速度的角度来说，这个新基金会如何解决这类问题？我们以前也遇到过，但程度没那么深，从来不像今天这样。
37:36
Yeah, we knew going in when we were standing in the foundation up, in fact, that was a question.
是的，我们一开始就知道，当我们建立这个基金会的时候，其实那是个问题。
37:42
Do we do we do a foundation foundations are slow.
我们要不要做一个基金会？基金会动作太慢了。
37:47
Yeah, I guess that's how people think.
对，我猜大家就是这么想的。
37:50
This is not slow. This is not a slow moving space.
这不慢。这不是一个缓慢发展的领域。
37:54
Do we do this, right?
我们这么做，对吧？
37:57
We realize we have to do this because you just have to have these neutral homes for things that are this critical but we cannot move at the pace that your traditional foundation will move at.
我们意识到必须这么做，因为对于如此关键的东西，你必须得有中立的归属，但我们不能按照传统基金会的节奏来推进。
38:11
So we move a lot faster and I think it really helps that the folks on a working group.
所以我们推进得快得多，而且我觉得工作组的成员们真的帮了大忙。
38:18
These are not volunteers doing this in their spare time.
他们不是利用业余时间来做这件事的志愿者。
38:21
Like it is part of their job, you know, to innovate the future.
这就像他们工作的一部分，你知道，去创新未来。
38:25
And if you know, hey, we we are trying to build a product on this standard.
而且如果你知道，嘿，我们，我们正试图基于这个standard来构建产品。
38:32
Then you're going to speed everybody else up.
那你就会让其他所有人都加速前进。
38:34
We have to come to some conclusions here, right?
我们总得在这里得出一些结论，对吧？
38:37
And I think that really helps is that the innovators themselves are the members of these working groups.
我觉得真正有帮助的是，创新者本身就是这些工作组的成员。
38:43
And so they have basically they they they have to move much faster.
所以他们基本上就是，他们必须得快得多。
38:50
They have an incentive to not drag this along, you know.
他们有动力不去拖延这件事，你知道的。
38:55
It does. I'm curious, you know, when you see when they come into the working group, is it, you know, and you're talking, you know, what are to the rest of us on the outside?
确实。我很好奇，你知道，当你看到他们进入工作组的时候——你在说——对我们这些外面的人来说，他们是什么样的？
39:05
Very fierce competitors, you know, you know, open AI and quad.
非常激烈的竞争对手，你知道，你知道，open AI 和 quad。
39:08
Yeah.
对。
39:09
You know, they're hammering out and Googles in there and you know, there's a whole slew of them.
你知道，他们在里面敲定方案，谷歌也在，你知道，还有一大堆。
39:14
And is the is the is the tenor of the conversation a little bit.
那么，就是，就是，对话的基调是不是有点……
39:21
I guess because they got to get stuff done.
我猜是因为他们得把事儿办了。
39:23
As you said, you got to get stuff done because everyone's waiting to move on.
就像你说的，你得把事儿办了，因为大家都在等着往前走。
39:28
Does that collaboration do they kind of just draw?
那种合作，他们是不是就是直接翻篇了？
39:30
I'm just just as a sheer curiosity.
我只是纯粹出于好奇。
39:32
They just drop kind of that external competitive behavior and just get in and say, yeah, let's just get it done.
他们就是放下那种对外竞争的心态，直接进来说，好，咱们就把这事儿干了。
39:37
And yes, they do.
对，确实如此。
39:38
It's fascinating.
这很有意思。
39:39
They actually are very friendly with each other, right?
其实他们彼此都非常友好，对吧？
39:43
So like sure, like, yeah, our company's just got into this heated Twitter war or something.
所以就像，当然，嗯，我们公司刚卷进了一场激烈的 Twitter 大战什么的。
39:49
But you and I got to feel this is standard off.
但你我都得觉得这是台面下的常态。
39:52
You know what I mean?
你懂我意思吧？
39:53
You can let those crazy people talk, but we got to get this thing done right now.
你可以让那些疯子在那边说，但我们得现在就把这事搞定。
39:57
Super collaborative, which is amazing.
超级协作，真是了不起。
39:59
It's amazing to watch.
看着真让人惊叹。
40:01
I tell people, like if you want to see the beauty of like open source and people working together from competing companies,
我跟人说，如果你想看到 open source 的美妙之处，以及来自竞争公司的人一起合作，
40:10
all you got to do is go in the MCP discord server.
你只需要进MCP的Discord服务器就行了。
40:13
It's a thing of beauty.
那真是一道美丽的风景。
40:14
There's just like dozens of these working groups within that protocol itself.
那个protocol本身就有几十个这样的working groups。
40:20
And you have members from everywhere, every company that's doing real work, not just throwing ideas out,
而且成员来自四面八方，每家真正在做事的公司，不是光抛想法出来的那种，
40:27
but doing real work to add new features or figure out how to do this other new thing or solve this problem.
而是真的在干实事，加新feature、研究怎么搞出别的新东西、或者解决某个问题。
40:37
Because their company's depends on those things, right?
因为他们的公司就靠这些吃饭，对吧？
40:41
And so it's beautiful.
所以这真的很美。
40:42
It's actually really beautiful to see.
看到这一幕确实特别美。
40:44
So as we start winding up here, I'm curious.
那在我们收尾的时候，我挺好奇的。
40:48
Like it's just kind of, you know, just the speed of operation is a little bit mind-boggling.
就是那种，你知道，整个运转的速度有点让人难以置信。
40:55
As you're looking ahead, and you guys have put together this foundation,
当你们展望未来，你们已经搭建好了这个基础，
40:58
and you're having those really productive and rapid conversations to get stuff done
而且你们正在进行那些非常有成效、快速的对话来把事情搞定，
41:05
so that this world continues at the velocity it's at.
好让这个世界继续保持它现在的速度往前跑。
41:09
Like, what are some of the things that you might expect to see?
那么，你觉得可能会看到哪些事情？
41:13
Having come along the path, because I think you're in a bit of a unique position to say,
一路走来，因为我觉得你处于一个挺独特的位置，可以说，
41:19
you know, you've built into this.
你知道，你已经深度参与进来了。
41:21
You're one of the people that set the foundation up.
你是奠定基礎的人之一。
41:24
You've seen it start to work, and you've seen it working well.
你看到它開始奏效，也看到它運行得很好。
41:27
And with those kinds of relationships forming and everything,
隨著這種種關係的形成，還有所有的一切，
41:33
how do you see things moving forward?
你覺得未來會怎麼發展？
41:36
You know what?
你知道嗎？
41:37
And you could be a little speculative.
你可以稍微推測一下。
41:39
It's fine to be wrong.
說錯沒關係。
41:40
I say things on the show all the time that are wrong.
我在節目上經常說錯話。
41:43
Daniel and I are like, oh, we got that one wrong.
Daniel和我就在想，哎呀，这个我们搞错了。
41:46
But we try.
但我们还是愿意尝试。
41:47
But it's fun to try to think, where might things go?
不过去想想事情可能会往哪走，也挺有意思的。
41:52
What are your thoughts?
你怎么看？
41:54
Like, where do you think?
就是说，你觉得呢？
41:55
Where do you think we're heading into this brave new world where everyone's trying to figure out?
你觉得我们会走向那个每个人都在摸索的勇敢新世界吗？
42:00
Not just like, like, no one knows 10 years, but like, like, people are trying to figure out,
不是说十年后没人知道，而是大家现在都在猜，
42:04
what's it going to be in six months?
六个月之后会是什么样？
42:06
Yeah.
嗯。
42:07
Yeah.
嗯。
42:08
What are your thoughts around that?
你对此怎么看？
42:09
If we look at, like, you and I, you know, and probably everyone listening to this podcast,
如果我们看看，比如你和我，你知道，可能还有收听这个播客的所有人，
42:14
we kind of live in this bubble where, you know, we're likely very exposed to AI.
我们都活在某种泡泡里，你知道，我们对AI的接触很可能非常多。
42:21
We're working with it daily.
我们每天都在跟它打交道。
42:23
That is not the case across the world.
但全世界并不是这样。
42:27
Like, you know, the general population, like you said, one, they hate AI.
就像，你知道，一般大众，就像你说的，第一，他们讨厌AI。
42:34
And two, they probably are not using it like in their day to day, right?
还有第二点，他们可能并不会在日常生活里用到它，对吧？
42:39
Maybe they've asked chat, GPT, a question, you know, they searched from something for Google
也许他们问过 ChatGPT 一个问题，你知道，他们在 Google 上搜了点东西
42:45
and a little summary came up.
然后出来一小段摘要。
42:47
And that's pretty much the extent of what they've done.
而他们做过的也就这么多了。
42:50
So outside of our bubble, like, we have so many challenges right now,
所以，在我们这个圈子之外，我们现在有太多挑战，
42:54
just amongst, like, the tech community.
哪怕只是在科技社区内部。
42:57
Outside of that, when you start looking at how, like, the everyday person is going to utilize agents
除此之外，当你开始思考普通人会怎么使用 agents，
43:06
for all sorts of things, right?
去做各种各样的事情，对吧？
43:09
Within their lives, you just start seeing all of these other things that we need to figure out.
在他们的有生之年，你会开始看到所有这些我们还需要搞清楚的事情。
43:14
And so I think that there's a lot we have to figure out.
所以我觉得有很多事我们得去弄清楚。
43:18
I think that'll consume us for the next couple of years.
我觉得这会在接下来几年里占据我们很多精力。
43:22
I do think that AI will become a part of, you know, the everyday person's day to day,
我确实认为AI会，你知道，成为普通人日常生活的一部分，
43:31
just like mobile phones and internet have, I think the same will exist.
就像手机和互联网已经做到的那样，我觉得AI也会这样。
43:37
I hope, I don't know, but I hope we're not just giving it all to the agents.
我希望，我不确定，但我希望我们不是把一切都交给agents。
43:43
And we become workers for the agents, you know what I mean?
然后我们变成给agents打工的人，你懂我意思吗？
43:48
I hope that we kind of get to that middle ground where we find,
我希望我们算是能达到一个中间地带，在那里我们发现，
43:53
even though it's capable, this is not a good use of it.
尽管它有能力，但这么用并不好。
43:57
And these are the ways that we should be deploying it.
而这些才是我们应该部署它的方式。
44:01
Fantastic. That's some, I share that with you.
太棒了。这点我跟你看法一样。
44:04
I think we need to find that, you know, where does the human fit into the equation
我觉得我们需要找到，你知道，人类在这一局面中的位置。
44:08
and where is AI a tremendous utility?
以及AI在哪些地方能发挥巨大作用？
44:13
Angie, thank you very much for coming on the show.
Angie，非常感谢你来做客我们的节目。
44:15
This was a great conversation.
这次对话非常棒。
44:17
Really appreciate it.
真的很感谢。
44:18
It gave me a lot to think about.
这让我想了很多。
44:20
I plan to dive into some of those projects myself and learn a bit more about them.
我打算自己深入研究其中一些项目，多了解一些。
44:25
And thanks for coming on the show.
感谢你来做客本期节目。
44:27
Hope to have you back sometime.
希望以后还能请你来。
44:28
Thanks so much, Chris. I enjoyed it.
非常感谢，Chris。我很喜欢这次聊天。
44:30
Alright, that's our show for this week.
好了，这就是我们本周的节目。
44:37
If you haven't checked out our website, head to practicalai.fm
如果你还没看过我们的网站，请访问 practicalai.fm
44:41
and be sure to connect with us on LinkedIn, X, or Blue Sky.
并且一定要在 LinkedIn、X 或 Blue Sky 上与我们联系。
44:44
You'll see us posting insights related to the latest AI developments.
你会看到我们发布关于最新AI发展的见解。
44:48
And we would love for you to join the conversation.
我们很希望你能加入对话。
44:51
Thanks to our partner Prediction Guard for providing operational support for the show.
感谢我们的合作伙伴Prediction Guard为节目提供运营支持。
44:55
Check them out at PredictionGuard.com.
可以去PredictionGuard.com看看。
44:58
Also, thanks to Breakmaster Cylinder for the Beats and to you for listening.
另外，感谢Breakmaster Cylinder提供Beats，也感谢你的收听。
45:02
That's all for now.
今天就到这里。
45:03
But you'll hear from us again next week.
但下周你会再次听到我们的声音。