桥水如何把一亿份文档转化为研究优势：Sam Green 与 Aaron Linsky

简要描述Brief Description

Sam Green 和 Aaron Linsky 介绍了 Bridgewater 如何构建 Vault，让逾 1 亿份非结构化文档为投资者和智能体所用。他们探讨了搜索质量、筛选与规模、投资者习惯的变化，以及能够捕捉并不断积累公司经验的主动型智能体。Sam Green and Aaron Linsky explain how Bridgewater built the Vault to make more than 100 million unstructured documents useful to investors and agents. They discuss search quality, filtering and scale, changing investor habits, and proactive agents that capture and compound the firm’s learning.

目录Table of Contents

• 简介：构建 Vault 的三点经验教训
  Intro: three lessons from building the Vault
• 桥水基金的50年研究流程
  Bridgewater's 50-year research process
• 为什么非结构化数据也需要同样严谨的处理方式
  Why unstructured data needed the same rigor
• 保险库里有什么
  What's in the Vault
• 开放网络与 Vault：并列对比
  Open web vs the Vault: a side-by-side
• 面向人类的搜索 vs 面向智能体的搜索
  Search for humans vs search for agents
• 精度基准测试：“Vault 搜索很糟糕”
  Benchmarking precision: "Vault search is bad"
• 大规模快速筛选
  Fast filtering at scale
• 让投资者改变他们的习惯
  Getting investors to change their habits
• Vault 的功能：分析、预测和信号
  What the Vault enables: analysts, forecasts, signals
• 主动式智能体：投资者工作台
  Proactive agents: the investor workbench
• Vault 如何累积公司所学的经验
  How the Vault compounds what the firm learns

引言：构建 Vault 的三点经验教训Intro: three lessons from building the Vault

Sam Green：大家早上好。我是 Sam Green，Bridgewater Associates 应用人工智能团队负责人，这位是我的同事 Aaron Linsky。今天我们来和大家聊聊 Vault——这是我们在 Bridgewater 构建的数据平台，旨在将非结构化数据转化为投资流程中的一项核心资产。Sam Green: Good morning everyone. I'm Sam Green, team lead for applied AI at Bridgewater Associates, and this is my colleague Aaron Linsky. And today we're here to talk to you about the Vault, the data platform that we've built at Bridgewater to turn unstructured data into a first-class asset in our investment process.

作为今天上午演讲的一部分，我们想分享构建 Vault 过程中学到的三条经验。第一，人类和智能体对数据平台的需求截然不同，因此必须分别针对这些不同需求进行构建。第二，当一个工具能够同时满足这两类需求时，才能产生最大的影响。And as part of the talk this morning, we'd like to share three lessons that we've learned building the Vault. The first is that humans and agents have very different needs from a data platform. And so you have to build for those different needs separately. The second is that the greatest impact occurs when a single tool can meet both of these sets of needs at the same time.

第三点是，真正的价值在于，你能够搭建起必要的基础，让智能体主动参与到你的投资流程中，或主动参与到你使用工具所做的任何事情中，而不是仅仅充当供人类使用的被动工具。And the third is that the real prize is when you can build the foundations needed to make agents a proactive participant in your investment process or whatever you're doing with your tools instead of a reactive tool for your humans to use.

桥水的50年研究流程Bridgewater's 50-year research process

在深入探讨 Vault 之前，我想先介绍一些 Bridgewater 的背景。过去 50 年来，Bridgewater 一直致力于尽可能深入地理解市场和经济如何运作，并将这种理解系统化地编码，以便随着时间推移对我们的逻辑进行压力测试。Before I dive deeper into talking about the Vault, I wanted to start with some background about Bridgewater. Over the last 50 years, Bridgewater has been in the constant pursuit of the world's best understanding of how markets and economies work. And turning that into a systematic encoding that we can use to stress test our logic over time.

最初，这一切只是写在黄色法律便笺纸上的一组“如果……那么……”规则，后来逐渐发展成一个持续复利的过程，以人为核心，不断寻找市场中下一个正在发生变化的事物。最近，我们一直在致力于打造 AiA，也就是人工智能投资助理，让机器成为这一持续过程的核心。在所有这些演进阶段中，我们始终秉持着一个理念：持续关注世界如何变化，以及世界上正在发生什么。And what originally began as a set of if-then rules on a yellow legal pad evolved into a continuous compounding process with humans at the center constantly looking for the next thing that's changing in markets. And more recently, we've been working on building AiA, the artificial investment associate, which puts a machine at the center of this continuous process. Across all these phases of evolution, we've had a constant philosophy in mind, which is that we want to be constantly looking at how the world is changing and what's going on in the world.

注意到哪些动态尚未被纳入我们的认知，并利用这些经验不断改进、积累优势，为未来打下基础。在桥水历史的大部分时期，这意味着构建并监督基于结构化数据的时间序列模型。而这些结构化数据必须经过精心构建，确保能安全地用于我们的研究流程，并与我们的内部本体和世界模型保持一致，这样我们才能充分发挥所构建数据的价值。Noticing what dynamics are not yet encoded in our understanding and use that learning to improve and compound for the future. For most of Bridgewater's history, that meant building and overseeing time series models on structured data. And that structured data needed to be carefully built so that it could be safe for use in our research process and aligned to our internal ontology and world model so that we could get great leverage out of using the data that we had built.

为什么非结构化数据也需要同样的严谨性Why unstructured data needed the same rigor

随着语言模型的出现，我们获得了一个全新的巨大机遇，得以接触范围广阔得多的数据。比如新闻和研究资料、文字记录、政策文件，以及此前只存在于无形之中的所有内部对话。也就是说，公司内部的备忘录、会议记录和各种讨论。我们很快意识到，所有这些数据都需要经过完全相同的处理，才能满足实际用途。With the emergence of language models, we had a huge new opportunity to engage with a much wider world of data. You know, news and research, transcripts, policy documents, and all of the internal conversation that was previously just in the ether. So, the internal memos and meeting transcripts and debate happening inside the firm. We quickly realized that all of that data needed the exact same treatment to be fit for purpose.

我们需要对它进行仔细清理和治理，并做好准备，以便既能用于当下的时点研究问题，也能支持我们长期开展系统性研究。纵观数据供应商和数据平台提供商，我们发现没有任何一家能提供我们所需的全部控制权，帮助我们建立并保持数据优势。It needed to be carefully cleaned, governed, and made ready for use in both point-in-time research questions today and also our systematic study over time. And when we looked out at the landscape of data vendors and data platform providers, there was nothing that provided us with all of the control that we needed to build and maintain our data advantage.

保险库里有什么What's in the Vault

因此，我们构建了 Vault，这是一个存储外部和内部非结构化数据的资料库，也是一个对所有数据进行索引和整理，并使其可供人类投资者和智能代理使用的平台。如今，Vault 中已有超过 1 亿份文档，每天每分钟都有数千份新文档加入，数据来自 135 个外部来源，而且来源数量还在不断增加；此外，Vault 还包含 Bridgewater 内部的所有数据。至关重要的是，这些数据并不是一股脑地堆到投资者的办公桌上。And so, we built the Vault, which is our repository of external and internal unstructured data. And the platform that indexes, organizes, and makes all of it ready for use by our human investors and agents. Today, the Vault has more than 100 million documents in it with thousands more added every day, minute by minute, from 135 external sources and again growing all the time, as well as all of the data from inside of Bridgewater. And crucially, all that data is not just dumped on our investors' desks.

它经过精心校准，与我们的内部本体保持一致，并补充了国家、概念、市场，尤其是文档中讨论的时间段等信息。It is carefully aligned to our internal ontology with enrichments like the countries, concepts, markets, and most importantly, time period discussed in the document.

开放网络与 Vault：并排对比Open web vs the Vault: a side-by-side

为了让这个例子更直观，我想先从一个简短的引入性示例开始。我向一个企业聊天客户端提出了一个相当基础、具有代表性的投资问题：“人工智能基础设施投资对韩元有何影响？市场参与者是否预计这种影响会持续下去？”大家都很熟悉这个过程：由于我明确要求它去寻找其他人对此问题的看法，智能体便会访问网络，查找可供参考的资料。To bring this to life, I want to start with a quick motivating example. So, what I did here is I asked a fairly basic representative investment question to an enterprise chat client. I asked, "How is AI infrastructure investment impacting the Korean won, and do market participants expect this to continue?" And as you're all familiar with, the agent goes to the web and looks for resources it can draw on because I've explicitly asked it to go find other people's perspectives on the question.

你可以看到，它会放大查看，并找到 15 个看起来相关的资源作为依据，然后才给出答案。我仔细看了它返回的答案后，有几件事让我印象深刻。首先，尽管它找到了 15 个看起来相关的来源，却只挑了一两个来回答我的问题。And you can see it zooms in and it finds 15 relevant-looking resources to draw on before it gives me back its answer. And when I zoomed in on the answer it gave back, a couple of things stood out to me. The first is that even though it found 15 relevant-looking sources, it only picked one or two to answer my question.

第二处在这条回复的末尾，智能体自己承认并未回答我的问题。它没能在开放网络上找到足够多的文档，无法据此确定市场共识，也无法满足我提出的条件。我再进一步仔细查看后，发现问题变得更严重了。And the second is down at the end of this response where the agent itself acknowledges that it hasn't answered my question. It didn't find enough documents on the open web to establish a market consensus and meet the parameters that I laid out. When I double-clicked even further, the problem got even worse.

它主要依据来回答我问题的其中一条笔记，早在2026年7月就已发布，尽管我是在上周末、也就是10月初才提出这个问题并记录下这个回答。在人工智能时代和投资领域，三个月的时间非常长。依靠这份资源来回答我的问题，远远不够。One of the notes that it primarily drew on to answer my question was published way back in July of 2026, even though I asked this question and captured this answer just at the end of last week in early October. In the age of AI and in investing, three months is a massive amount of time. It's not even close to good enough to draw on this resource to answer my question.

当我在关联了知识库的情况下提出完全相同的问题时，结果却截然不同。首先，连接到知识库有助于提示智能体：回答我的问题时，应该查找特定时间段的信息。When I ask this exact same query attached to the vault, something totally different happens. The first is that the connection to the vault helps to prompt the agent that it should look for a specific period of time when answering my question.

因此，它将范围缩小到 10 月 1 日至 10 月 8 日。由于资料库中与投资问题相关的内容语料丰富得多，它找到了范围广泛得多的资源：在这段特定时间内，有 52 份候选文档与查询相关。最终回答我的问题时，它引用了 7 份文档，而不是仅仅一两份。So, it zooms in on October 1st to October 8th. It finds a much wider array of resources, 52 candidate documents that are relevant to the query over this specific period of time because of the vault's much more significant corpus of content that's relevant to investing questions. And when it ultimately answers my question, it draws on seven instead of only one or two.

并给我一个丰富的答案，让智能体更有把握地认为它能够解答我的问题。因此，对我们来说，构建这个知识库带来的重要启示，也是今天这场分享的主题，就是要让智能体发挥影响力，离不开优质数据和合适的连接器。接下来，我把时间交给 Aaron，请他谈谈构建这个知识库过程中在产品工程方面的一些经验。And gives me back a rich answer that the agent is more confident is going to address my question. And so, for us, this is the big lesson from building the vault to kind of motivate the talk today, which is that it takes great data and the right connectors to make agents impactful. And I'm going to hand it over to Aaron now to talk about some of the product engineering takeaways from building the vault.

面向人类的搜索 vs 面向智能体的搜索Search for humans vs search for agents

Aaron Linsky：谢谢你，Sam。大家好，我是 Aaron，在 Bridgewater 负责产品工程。接下来，我会分享过去几年里我们构建 vault 的一些经验。我想先谈一条看似显而易见、但我仍觉得值得一提的经验：搜索很难。Aaron Linsky: Thanks, Sam. So, hi. I'm Aaron. I run product engineering at Bridgewater. I'm going to take us through a few of our learnings from building the vault over the past few years. I'm going to start with one that might seem obvious, but I still think it's worth saying, which is search is hard.

但是，如果我们要为两类不同的使用者设计，事情就会变得更难。而在这里，我们要为人类和智能体设计。接下来，我会先从人类说起，谈谈两者之间的区别。人类需要精确的信息和较高的信噪比，因为我们的阅读能力有限。即使是阅读速度很快的人，每分钟也只能读大约 300 个单词，而我个人读得慢得多。But, it gets even harder if we're designing for two different consumers. And in this case, we're designing for humans and agents. I'm going to go over the difference here starting with us. So, humans need precision and a high signal-to-noise ratio because we can only read so much. Even a fast reader manages about 300 words a minute, and personally, I'm much slower.

智能体是一种不同类型的读者。它们对精确性的敏感度要低得多。只要数据在那里，它们就能找到。基本上，你交给它们的任何内容都会得到处理。Agents are a different kind of reader. They're far less sensitive to precision. If the data is there, they'll find it. And basically, anything you hand to them will get attended to.

它们的规模增长非常快，消耗速度也快了大约一百倍，达到每分钟 30,000 个单词。因此，我们需要为这两者构建 Vault。谈到重要的维度，白板上列出了五项：召回率、精确率、过滤、权限和规模。其中一些对人类用户很重要，一些对智能体更重要，还有许多对两者都很重要。接下来我会深入探讨其中三项：对人类用户而言的精确率、对智能体而言的规模，以及对两者都重要的过滤。They scale super quickly and they consume about a hundred times faster, 30,000 words per minute. And so we need to build Vault for both of these. And when we're thinking about the dimensions that matter, there are these five on the board. Recall, precision, filtering, entitlements, and scale. Some of them matter to people, some matter more to agents, many to both. I'm going to go deep on three of these. Precision for people, scale for agents, and then filtering for both.

基准测试精度：“Vault 搜索很糟糕”Benchmarking precision: "Vault search is bad"

说到精确率。我们填满 Vault 后——这可是数据团队和法务团队的一项艰巨任务——召回率就不再是亟待解决的问题了。不过，大家最常抱怨的是 Vault 搜索很糟糕。作为产品负责人，听到这种评价很难受，但我的回应是：那我们就来做基准测试。于是我们就这么做了。So, precision. Once we'd filled the Vault, which was a heroic effort by our data and legal teams, recall was no longer a pressing issue. However, the number one complaint was Vault search is bad. As a product owner, it's painful to hear that, but my response was let's benchmark it. And that's what we did.

我们收集了一些问题，并手动找出了最合适的文档，其中也包括那些看起来有帮助、实际上却没有帮助的文档。总共大约有 200 个样例，耗时不到一周。这件事既不有趣，也不怎么光鲜，但非常有帮助。评估之后，结果非常明确：搜索效果很差。我们关注的关键指标是前 10 条结果的精确率，也就是前 10 条结果中有多少是正确的文档。We collected questions and found the best documents by hand, including ones that looked helpful but weren't. In total it was about 200 examples and it took us less than a week. It was not fun or very glamorous, but it was very helpful. And after we ran it, it was very clear. Search was bad. The key metric we were looking at is precision at 10, which is of the top 10 results, how many are the right documents.

而我们的命中率大约是 60%。所以，人们说得没错，情况确实很糟。And we were hitting about 60%. So, the people were right, it was bad.

但现在，我们有办法衡量所有这些改进了。白板上列出了其中五项。等到我们完成 Agentic Search 时，Precision@10 已经达到 90%–95%。好，这就是精确率。But now we had a way to measure all of the improvements. And here on the board are five of them. And by the time we got done with Agentic Search, we were up in the 90–95% precision at 10. All right, so that was precision.

大规模快速筛选Fast filtering at scale

接下来谈谈筛选。和任何人一样，我们都希望找到最相关的文档。但查看生产环境的跟踪记录时，我们发现，几乎总是会有某种筛选条件，比如截至某个日期、来自某个来源，或涉及某项特定资产。所有这些都需要高级筛选，而这正是我们遇到的第一个重大性能问题。Now on to filtering. Like anyone, we want the most relevant documents. But when we looked at the production traces, what we were seeing is that almost always there was some condition, either as of a certain date or from a certain source or about a specific asset. And all of that requires advanced filtering, and that's where we ran into our first major performance issue.

传统上，过滤向量数据库有两种方式。后置过滤会先使用向量索引，然后根据元数据对排名靠前的结果进行过滤。这种方式速度快，但召回率可能很低，因为全局相关的结果可能不符合你的过滤条件。另一方面，预过滤会先进行过滤，然后计算所有匹配项之间的距离。There are classically two ways to filter a vector database. Post filtering will use that vector index first and then filter the top results by metadata. It's fast, but recall could be pretty low because the globally relevant results might not match your filters. On the other side, we have pre-filtering where we'll filter first and then calculate distances for every one of those matches.

这种方法的优势是召回率很高，但速度可能非常慢，而我们需要一种介于两者之间的方案。因此，我们开始尝试 Turbo Puffer，并最终转而采用它。他们将这种方法称为原生过滤：只在包含匹配项的最近簇中查找，然后仅对这些匹配项评分。The advantage there is that the recall is very good, but it can be very slow, and we needed something in between. So that's what led us to try and then ultimately switch to Turbo Puffer. Their approach, which they called native filtering, looks only in the nearest clusters that contain matches and then scores just those.

最终得到的方案，速度和后置过滤一样快，召回率却接近预过滤。最后，我要谈谈规模。对我们来说，人工使用规模相当普通：每小时 100 次查询。我们有 300 名用户，他们会时不时地进行搜索。用户需要几分钟才能读完收到的文档。What you end up getting is something that's fast like post filtering, but with close to the same recall as pre-filtering. And lastly, I'm going to talk about scale. Human scale for us is pretty pedestrian, 100 queries an hour. We have 300 users. They're going to search, you know, every so often. It takes our users minutes to read the docs that they're receiving.

而智能体则比这高出几个数量级。每秒轻松处理 100 多次查询，智能体集群更是轻易就能达到每秒数千次。Agents, on the other hand, are a few orders of magnitude beyond that. Easily 100 plus queries a second, and agent swarms can hit thousands easily.

而且显然，正如我之前所说，他们会在几秒钟内读完文档。所以我们也需要一个能够扩展的方案，而 Turbo Puffer 也轻松应对了这一点。因此，我们涵盖了全部五个维度。我还没谈到权限权益，不过今天晚些时候会讲到。但我们把所有方面都涵盖了。它解锁了新的工作流。And they obviously, will read, as I said earlier, will read docs in seconds. So we needed something that could scale as well, and again, Turbo Puffer handled that easily, too. So we were covering all five dimensions. I didn't talk about entitlements, but I'll be talking about that later today. But we covered it all. It unlocked new workflows.

促使投资者改变习惯Getting investors to change their habits

但这并没有改变研究人员的工作方式。他们早上还是不会登录 Vault。这就是我想在这里分享的第二个经验：无论工具多么出色，我们都不能只是把它们放到投资人的办公桌上，就指望他们会用。我们需要把他们的旧习惯迁移到新的形式中。于是，我们研究了他们原本的做法。很多人都依赖电子邮件。研究资料直接来自经纪商。他们订阅了各种邮件列表，还设置了规则，把所有内容筛选到不同文件夹里。这套做法行之有效，也让他们感到熟悉，而且他们知道自己不会漏掉任何信息。因此，我们必须在 Vault 中实现所有这些功能。我们正是这么做的。所有内容都可以搜索。订阅列表变成了已保存的搜索，文件夹则变成了已保存的收藏集。But it didn't change how our researchers worked. They still weren't coming to Vault every morning. And that's my second lesson that I want to share here is that no matter how amazing the tools are, we can't just drop them on our investors' desks and expect them to be used. We need to migrate their old habits into a new form. And so we looked at what they were already doing. Many of them were living in email. The research came straight from the brokers. They had subscribed lists. They had rules that filtered everything into folders. It worked. It was familiar. And they knew that they weren't missing anything. So we had to build all of these features into the Vault. That's exactly what we did. Everything's searchable. The subscribe lists became saved searches. And folders became saved collections.

习惯还是那些习惯，只是换了个新家。一旦把它们带到那里，我们就能开启新的使用习惯。他们可以像 Sam 展示的那样，针对所有数据提问；可以按主题关注动态信息流；还可以把这些内容集合交给自己的智能体，采取进一步行动。就在那时，我们清楚地找到了产品与市场的契合点。找到契合点后，用户希望到处都能用上它。于是我们把它集成进聊天工具，开发了手机应用，也让它出现在 Slack 和他们的 IDE 里。当然，我们也让智能体能使用所有这些功能。其中还包括 CLI，因为智能体很喜欢 CLI。It's the same habits, just a new home. And once we had them there, we can open up new habits. So they can ask questions across all the data as Sam showed. They can follow dynamic feeds by topic. And they can hand those collections over to their agents to take further action. And that's when we clearly had product market fit. And once we had that fit, the users wanted it everywhere. So we put it in chat. We made a phone app. It's in Slack in their IDE. And of course, we availed all of that to the agents as well. Including a CLI because agents love CLIs.

Vault 的功能：分析师、预测、信号What the Vault enables: analysts, forecasts, signals

所以这里的核心问题是：好吧，这能带来什么？Sam 接下来会介绍我们目前正在打造的产品。但首先，也是最重要的，它能帮助我们更快地完成研究闭环。借助我们的“口袋分析师”——也就是首个 AI 分析师——我们只需花几美元，就能在几分钟内完成原本需要数小时的研究。And so the main question here is okay, so what does this enable? Sam's going to talk about what we're building right now. But first and foremost, it's helping us to move around that research circle faster. With our pocket analyst, which is our first AI analyst, it can do hours of research in just a few minutes for a few dollars.

这依赖于 Vault。现在，我们正借助 Researcher 进一步拓展，这代表着下一代产品。它的理念相同，只是将时间跨度延伸到了几天或几周。我们还能够训练定制模型，让它们更智能地对 Vault 中的数据进行推理，并将这些模型与其他基础模型结合起来，为系统化的决策流程提供支持。And that's relying on the Vault. And then we're expanding now with the researcher, which is the next generation. It's the same idea stretched across days or weeks. We're also able to train custom models to more intelligently reason over the vault's data and use those together with other foundation models to feed systematic decision-making processes.

预测器是我们的首个产品，可生成与宏观经济相关的 AI 预测。此外，我们还基于 Vault 构建了投资信号，将非结构化数据转化为结构化、可交易的策略。The forecaster is our first one, which produces AI-generated macro-relevant predictions. And then, we have also built investment signals from the vault, turning unstructured data into structured, tradable strategies.

现在，我把时间交给 Sam，请他为我们介绍 Vault 的最新功能。Now, I'll hand it to Sam to take us through Vault's latest capability.

主动型智能体：投资者工作台Proactive agents: the investor workbench

Sam Green：谢谢。就像我在演讲开头提到的，真正的价值在于：你可以让智能体不再只是一个被动的资源，随时准备回答人类的问题，而是成为主动的参与者，帮助引导你的高价值专业人士，将注意力集中到最重要的领域。我们通过一种称为“投资者工作台”的体验来解决这个问题。在这里，智能体始终处于运行状态，跟踪对投资者而言重要的事项，并向他们展示最需要关注的领域。Sam Green: Thanks. So, like I alluded to at the beginning of the talk, the real prize is when you can take agents from a reactive resource that can be ready to answer a human's question and instead make them proactive participants helping to steer your high-value professionals towards the most important areas for their attention. And the way that we're solving that problem is in an experience we're calling the investor workbench, where the agent is always on tracking what's important to that investor and presenting them with the areas that are most in need of their attention.

这看起来像是一种依赖知识库的体验：系统会提出一个主题，并向投资者展示若干信息，供他们判断是否要进一步深入了解，还是完全忽略该主题。It looks like an experience like this that relies on the vault, where a topic is proposed and several pieces of information are presented to the investor to decide whether they should dig deeper or ignore that topic entirely.

你可以在这里看到投资者会看到的卡片，投资者可据此决定是否要进一步深入了解我们在演讲前面讨论过的同一个问题：人工智能对货币的影响。投资者会获得所需的一切信息：如果想快速浏览智能体的发现，可以查看智能体生成的摘要；他们还可以看到智能体所依据的确切证据，从而自行判断是什么促成了这项建议。You can see here the card that an investor sees in order to decide whether to dig in deeper into the same AI impact on currencies question that we looked at earlier in the talk. And the investor receives everything they need, an agentic summary if they want a quick scan across the agent's findings. They also get the exact evidence that the agent was drawing on, so they can assess for themselves what was leading to this recommendation.

如果他们选择进一步深入，就可以直接获取 Aaron 刚才介绍的分析工具所生成的研究成果。而当他们与智能体一起深入探究驱动这个问题的各种动态时，所获得的见解以及协作成果都会自动记录并存入知识库。And if they choose to dig deeper, they have a direct connection to the research output from the analytical tools that Aaron was just describing. And when they dig deeper alongside the agent to look at the dynamics driving this question, the learnings are automatically captured, the results of their collaboration are automatically captured back in the vault.

Vault 如何复合积累公司的所学成果How the Vault compounds what the firm learns

因此，这对我们来说是关键经验。这个知识库能够实现复利效应，因为它能为投资者和代理人提供恰当的证据，以及触手可及的额外内部和外部信息；它还会自动捕捉他们形成的综合分析，并将其放回语料库，供下一轮使用。这样一来，我们就能更快、更好地循环往复。And so this for us is the key lesson. The vault enables compounding because it provides the right evidence, the right extra internal and external information at the fingertips of our investors and agents, and it automatically captures the synthesis that they generate and puts it back in the corpus for the next turn. And that makes us go around the circle faster and better.

因此，当我们回想起最初提出的问题时，要让智能体发挥作用，关键在于为它提供恰当的上下文，并让它正确理解你所在世界正在发生什么。那么，我们想问你的是：你的 AI 知道这个世界正在发生什么吗？它记得你的公司对此有哪些了解吗？And so when we think back to the question we began with, what it takes to make an agent work is the right context and the right understanding of what's going on in your world. And so our question for you is, does your AI know what's going on in the world? And does it remember what your firm has learned about it?

因为我们正在训练自己的模型。如果这听起来像一项令人兴奋的挑战，我们很乐意聊聊。Because we are teaching ours to. And if this sounds like an exciting challenge, we'd love to talk.

谢谢。Thank you.