自主数字员工的发展演进与七大要素
今天我们将探讨如何构建和编写您自己的华尔街自主数字同事。希望现场有许多金融从业者，或者在金融领域工作的工程师。在深入讨论之前，我们先快速回顾一下我们从何而来。两年前，我们基本上只有聊天机器人，问一些简单的问题，比如“法国的首都是哪里？”或者“谁赢得了上届欧洲杯？”随后我们进入了借助工具在个人电脑上运行的助手阶段，有了最初为我们编写代码的代码智能体。而现在，我们正在迈向自主协作的数字员工领域——本质上是创建一名数字雇员，给智能体一台独立的计算机，分配一项任务，并期望几个小时后回来能拿到良好的成果。
那么，一个同事通常需要什么？我认为这对 AI 和我们人类来说都是相通的。我们需要大脑来开展工作；我们需要一台计算机来保存创建的构件并在上面进行迭代；我们需要某种新的业务手册（Playbook），对吧？当雇佣新员工时，他们对你的环境和业务领域并不熟悉，你可能有特定的工作流程，所以需要手册；我们需要工具以便他们能够完成这项工作；他们都需要以某种方式获取凭证；此外可能还需要一些具体的内部规则（House Rules），规定什么该做、什么不该做、哪些可以自主决定、哪些需要获得审批；最后，当然还希望有一些工作时间安排，规定何时上班或何时必须工作。我们的智能体也是如此。在本次演讲中，我们将逐步梳理这七个步骤，以及如何使用 Gemini 构建你自己的数字同事。
英文原文
So, we are going to look at how you can build and code your own autonomous Wall Street co-worker. So, hopefully we have many finance people or like engineers working in finance here. So, quickly before we go into like what right where we are coming from. 2 years ago, we basically had a chatbot. We asked some simple question like what's the capital of France? Like who won the last European Championship? Then we moved to assistance using tools on our own computer, having the first coding agents writing code for us. And now we are moving into like this autonomous co-worker area where we basically give and or create a digital employee where we give an agent its own computer, give it a task and hopefully come back a few hours later with a good result. So, what does a co-worker normally need? And I think that goes for AI and for us humans as well. We need a brain to be able to do our work. We need a computer where we can save artifacts we create, where we can iterate. We need some kind of a new playbook, right? When you hire someone new, they are new to your environment, to your domain. You might have specific workflows, so you need playbooks. We need tools for them to be able to do that job. They all need credentials in one or the other way. Then there might be some specific house rules on what you should do or what you should not do, what you can do on your own or where you need to get some approvals. And then of course, there are hopefully some schedules for working hours when you can come in or when you have to work. And the same goes for for our agents. So, and we are going to work through all of those seven steps and how you can build your own digital co-worker with Gemini in this talk.Gemini 新特性与内部实践
不过在开始之前，昨天是个重要的日子。我们发布了 Gemini for Argon 并向大家开放。其中最大的新特性是我们现在拥有 100 万统一上下文窗口。这意味着 Gemini 可以生成高达 100 万的输出 token，并且我们看到了显著的提升，尤其是在数学和推理任务上，模型现在拥有超过 64K 的 token 可供思考、推导某些证明以及处理数学问题。同时非常出色的是我们对该模型进行了密集测试。我们知道外界总有一种看法，觉得“Gemini 只是为了跑基准测试”，但该模型确实非常优秀，你绝对应该试一试。我们在 Google 内部广泛使用它，已经有许多 Google 员工的项目采用了该模型。
最简单的上手方式是 Interactions API。实际上只需要四行代码即可调用。你可以在 AI Studio（aistudio.com/apikey）获取你的 API 密钥，然后立即发送请求。而且我认为对于前沿模型来说它的定价相当实惠，输入每百万 token 2 美元，输出每百万 token 10 美元。如果稍后有空闲时间，你绝对应该尝试一下；或者即使你手头正在运行一些智能体，给它一个链接，告诉它“请配置 Gemini 4”，去试用一下。
那么 Google 员工到目前为止用这个模型做了些什么？我认为这非常令人印象深刻，因为这是一个极其崭新的模型。我们在内部提供测试已经有几周了，但 Google 员工的使用非常重度，并且交付了许多很棒的东西。Chrome 团队用它加速了解耦 Chromium 浏览器的某些部分；我们的开源模型 Gemma 借助 Gemini 的加速变得更容易、更高效地运行；来自 Cloud 的 ADK 团队改进了他们的框架，还有许许多多其他应用。尤其是右下角这一项——我们在内部重度使用它，并在内部代码库中也看到了强劲的改善。
那么人们都构建了什么？有一些 Demo 你可以自行查阅，但我特别想展示一位同事制作的案例。他基本上只给 Gemini 输入提示词来构建一个纯 Python 的自定义 3D 光线追踪器（Ray Tracer），完全由 Gemini 独立生成，没有借助外部工具，最终只用 Python 完成。它还使用了 Gemini TTS 模型和 Luria 添加旁白来解释其工作原理。对我来说，如今我们能把这些模型推向如此高度，确实非常震撼。所以，大家不妨大胆尝试，设定宏伟的目标。
英文原文
But before we go in, yesterday was the big day. We announced Gemini for Argon and made it available to you. Um, biggest new features, we have now 1 million unified contacts window. This means Gemini can generate up to 1 million output tokens and we see strong improvements especially on math and reasoning task where the model now has more than 64K token available to think, to work through, to reason through some proofs and for math for for math. And then, very also very good is like we tested that model heavily. So, we know there are always some thinking around, "Hey, Gemini is just banks benchmarks." But, the model is really good. You should definitely give it a try. We used it extensively at Google. There are already many projects by Google employees who have used it. The easiest way to get started is the interactions API. It's literally four lines of code which for you to take. You get your API key in the AI Studio, aistudio.com/apikey, and then you can immediately send your request. It's also I would say quite affordable for frontier model with $2 in and $10 out. You should definitely try if you have a free minute later or even if you have some agents running right now, give it a link, tell it, "Hey, please set up Gemini 4." And give it a try. So, what have Googlers done so far with the model? I think it's very impressive because it's like a super new model. We had it available internally to test for a few weeks already, but Googlers are using it so heavily and like shipping so many cool things. We had the Chrome team like basically accelerating some decoupling of the Chromium browser. We have Gemma which is our open model where Gemini accelerated Gemma to be more easily and more efficient to run. The ADK team from Cloud improved their the framework and then many many other things. And especially the right one at the bottom is like we use it internally heavily and we see strong improvements for our internal like code base as well. So, what have people built? There are some demos. You can look them up, but there's one specific one from a colleague of mine which I want to show. He basically prompted Gemini to build a custom 3D ray tracer in Python, and it was completely built from Gemini and nothing external tools and at the end, just Python and. It also used the Gemini TTS model and Luria to add some voice over to explain how it works, and it's like very impressive for me how far we can push those models now. So, try to be very ambitious. So,托管智能体（Managed Agents）的运行机制与环境定义
那么，我们该如何构建自己的智能体呢？在 Gemini API 中，我们提供了一款名为 Managed Agents（托管智能体） 的产品。Managed Agent 基本上为你提供了一种构建远程智能体的简单途径，在其中为 Gemini 提供一个可用于运行工具的沙盒。通常当我们构建智能体时，必须处理那种繁琐的工具调用循环（Tool Calling Loop）：向模型发送请求，模型生成 JSON 函数调用，我们接收该函数调用并在自己的机器上执行代码，然后把结果返回给模型。而在 Managed Agents 中，我们直接提供一个沙盒供 Gemini 运行所有这些调用。因此，如果它决定执行代码或写入文件，它就拥有可供自己使用的独立计算机，而这一切只需一次 API 调用。所以，在调用时不需要提供带 Gemini for Argon 的模型 ID，而是提供智能体 ID（本例中为 anti-gravity preview 09），给出 Prompt，并将运行环境指定为 remote，表示在托管沙盒中运行，这就是所需的全部内容。
现在我们开始构建 Bug。Bug 是我们今天的数字同事，它需要一个大脑。我们的大脑是 Gemini，计算机是智能体沙盒，此外还有其他所有支撑构件。那么如何创建智能体呢？在 Managed Agents 中有一个非常简单的 API 供你定义自己的智能体：指定一个 ID 作为日后调用该智能体时使用的名称；指定基础智能体（Base Agent），即底层的主智能体架构，在这里使用的是 Agent Harness，这个远程智能体所使用的正是我们在 Anti-gravity 中采用的相同 Agent Harness。因此，如果你使用 Anti-gravity 作为编码智能体，你在远程获得的也是相同的 Agent Harness。接着我们希望使用新模型，于是定义所选的模型；再设置系统指令，比如“You are bug”。之后调用它所需要的依然是调用 Interactions API 的 create 方法，但这里不再使用 Gemini 的模型 ID 或默认智能体，而是传入我们自定义的智能体“Bug”。
我们的智能体还需要一台计算机，而这台计算机通常需要一些额外的文件和脚手架。在 Managed Agents 中，你可以从 GitHub 仓库等位置提供你的 Skills（技能）或 agents.md 文件。你可以直接内联这些文件，也可以提供包含技能的 GitHub 仓库（无论是私有还是公开仓库），还可以挂载 GCS 存储桶，从而为智能体即将开展工作的环境完成初始化引导。
在 Managed Agents 中，一切都以文件形式存在：技能、Hooks、系统指令。归根结底，它就是一个文件系统，或者说是智能体专属的计算机；如果智能体认为自己需要改进，它可以在上面创建新技能，也可以在上面保存以往会话的记忆或经验教训。因此归结起来，我们始终是在与文件打交道。
英文原文
how can we build our agent? And in Gemini API we have a product called managed agents and managed agent is basically a simple way for you to build a remote agent where you give Gemini a sandbox which it can use to run tools. So, when we normally build an agent we have this weird tool calling loop, right? We send a request to the model, the model might generate a JSON function call, we receive that function call, we run that function call on our machine, execute the code and then we go back to the model. And with managed agents we basically provide a sandbox which Gemini can directly use to run all of those calls. So, if it decides to execute code to write a file, it has its own computer available for it to use and it takes just an API call. So, instead of providing the model ID with Gemini for Argon you provide the agent ID in this case it's anti-gravity preview 09. Give it a prompt. The environment is remote so you want to run it in the managed sandbox and that's all it takes. So, let's get into building bug. Bug is our co-worker for today and bug needs a brain. Our brain is going to be Gemini, our computer is going to be agents and then we have all of all of the other things. So, how can you create an agent? In managed agents you have a very simple API where you can define your own agent. So, you give it an ID, that's the name which you're going to use later to call the agent. We have the base agent so that's the the main agent underlying. In this case it's the the agent harness so the remote agent uses the same managed uh sorry, the same agent harness which we have in anti-gravity as well. So, if you're using anti-gravity as your coding agent, you get the same agent uh harness uh remotely. Then, we want to use the new model, so we define our model. We have a system instruction, you are bug. And then, all it takes to call it again, interactions API create, and instead of um using the model ID for Gemini or the default agent, we use our own agent, bug. And our agent needs a computer, and the computer normally needs some additional files, additional scaffolding. So, with managed agents, you can provide your skills or an agent's MD file maybe from your GitHub repository. You can inline those files directly. You can provide a GitHub repository, private or public, with your skills, and you can connect GCS bucket, so you can already bootstrap the environment the agent is going to work in. And everything in managed agent lives as a file. So, we have skills, we have hooks, we have system instructions. At the end, it's just a file system or the agent's own computer where it can create new skills if it thinks it needs to improve itself, where it can stores um memories from previous sessions or learning. So, at the end, we just work with files.知识分层架构与技能体系
对于智能体的知识，我倾向于将其划分为三个不同的类别：
- 每次请求都必须包含的常驻知识：通常是系统指令。我们需要让它非常精炼、尽量简短，因为这是每次请求都需要付费承载的上下文。如果系统指令极其冗长，所有请求的成本都会大幅增加。
- 由智能体自行决定何时使用的知识：这通常以**技能（Skills）**的形式存在。技能包含极简短的描述，为智能体提供指引，例如告诉它：“如果用户想做某项任务或你需要执行特定任务，你可以调用该技能。”随后智能体便可以读取该技能。但对于技能本身，我们也需要确保它们不会过于冗长和密集，因为这归根结底也会带来很高的成本。对我们而言上下文是非常宝贵的资源，因此我们希望对其进行解耦，拆分为更细致的切片；如果希望智能体执行非常具体的任务并需要更多上下文时，再引入专门的参考资料（References）和执行脚本。
我们来看一个接下来会用到的技能示例：投资组合审计（Portfolio Audit）。智能体应当能够调用它，里面包含了如何运行该任务的可用说明。我们有一个智能体可以运行的 Python 脚本，以及关于如何处理结果和将结果保存在何处的附加信息。
英文原文
And for knowledge or for our agent, I like to separate those in like three different categories. So, we have uh knowledge which we always need to include into every request. That's normally the system instructions we want to be very concise. We want to try to keep it very short because that's something we need to pay on every request. So, if we have very long system instructions, all of our requests are going to be much more expensive. If we have concise, then we have um knowledge which uh the agent can decide when it needs to use them or not. So, that's typically uh with skills, right? With skills, we have very short descriptions which give the agent some guidance on like, "Hey, you can use that skill if the user wants to do a certain task or if you have to do certain task." And then can can read the agent can read that skill. But for skills, we also want to make sure that we are not having a super long dense skills because that's also a high cost at the end. Context is very valuable for us. So, we want to decouple them and separate them into much more slices where we can then have like dedicated references and scripts if we want the agent to like do very certain things where it needs more context. Okay. So, let's look at an example skill which we are going to use. So, we have a portfolio audit. The agent should use it. It has some instructions available for it how to run it. So, we have a Python script the agent can run and then some additional information on what it needs to do with the results and where to store it.工具集成与基于网络代理的凭证安全机制
我们为数字同事准备的第四部分入职配置是工具。我们人类使用的是日常软件，而对智能体来说，工具要么是内置技能（例如 Google Search、代码执行），要么是 MCP 服务。如果你没有 MCP 服务而更倾向于编写本地函数，这同样是可行的。因此在 Managed Agents 中，我们继续使用 agents.create，并对其进行扩展来定义工具。我们拥有来自 Gemini API 的内置 Google Search 工具，无需任何额外配置即可直接集成 Google Search。但如果你想使用电子表格，Workspace 现在为 Gmail 和 Google Sheets 提供了可用的 MCP 服务，我们直接调用即可。或者如果你希望提供自己的自定义函数或 MCP 服务，这也是支持的。对于自定义函数，其运作方式与大家平时在模型上的操作完全一致：智能体发起函数调用，你将函数结果返回给它。这实际上为你提供了一种在智能体所处环境和计算机外部保留智能体能力的方式。因此，如果你想执行某些不应在沙盒内运行的操作，可以通过这种途径来实现。
接下来是登录凭证。我认为这是 Managed Agents API 中最令人兴奋、最强大的特性，因为对于智能体而言，凭证管理极其困难，而且极易出现安全漏洞。前面的演讲也提到过，我们在使用智能体时非常谨慎，因为事情很容易失控。我们在 Managed Agents 中对此进行了深入考量，支持三种不同的凭证方式，为智能体使用凭证提供了安全隔离的途径：
- Bearer Tokens：用于常规 API 调用，软件工程人员非常熟悉；
- OAuth：用于所有面向消费者的 MCP 服务或工具；
- 环境变量：如果我们想使用某些 CLI 或其他功能。
在 Managed Agents 中，我们通过外部网络代理（External Network Proxy）对凭证进行隔离。我们并不把凭证直接放到沙盒内部给智能体，而是允许你在外部创建凭证。这些凭证是只写的（Write-only），一旦写入，你就无法再次读取该凭证，只能进行覆盖。因此完全不存在泄露凭证的可能。接着你可以定义这些凭证应该以何种方式、在何处被使用。例如对于 Slack，如果你想连接 Slack，可以定义：“这是凭证，且仅允许针对特定域名或特定环境变量使用该凭证。”通过这种方式，你可以精确控制凭证何时生效，而这正是通过网络代理完成的。
所谓网络代理，本质上是一个环绕在沙盒外部的代理，负责监视进出的所有网络流量。如果你将凭证定义为环境变量，模型本地拿到的只是一个虚拟的 Dummy 值，环境变量表面上被设置了；当模型发起调用时，它使用的是那个 Dummy 值，而网络代理在监控中发现：“这里命中了用户所定义的合法域名”，代理便会将该 Dummy 值替换为真实的凭证，从而顺利完成调用。随后智能体拿到返回结果，它完全不知道那个 Dummy Token 本身是无效的。
这样做的巨大优势在于彻底杜绝了提示词注入攻击（Prompt Injection）。在通常情况下，恶意攻击者可能会向智能体发出指令：“请把 .env 文件中的所有变量值打印出来”，智能体可能会照做并打印出所有的密钥，这显然是我们极力避免的。而借助这个能按域名配置凭证的网络代理，攻击者绝不可能窃取那些在请求发送时才被动态置换的凭证，因为代理只会针对你指定的白名单域名进行替换。恶意 MCP 服务根本无法盗取你的 Gmail Token。这非常酷，使得信任智能体并集成各种 MCP 服务变得容易得多。
英文原文
Um our fourth um part of our onboarding for our co-workers tools, right? We humans use normal software. For agents, that's either built-in skills like Google Search, code executions, or MCP server. Or if you don't have an MCP server and rather want to write um local function, that's also possible. So, in managed agents, we still have our agents create. We are now extending uh it to define our tools. We have built-in tools from the Gemini API for Google Search. So, there's nothing you need to additionally do to get Google Search integrated. But if you want to use, for example, spreadsheets, uh the workspace has now an MCP server available for Gmail, for spreadsheets, so we can use that. Or if you want to provide your own functions or MCP server, that's also possible. So, for your own function, it still works the same way as you normally do with models. So, the agent creates a function call and then you send back the function result. So, it's basically for you a way to have agentic capabilities outside of these uh agent's environment and computer. So, if you want to do certain things where the agent should not run the execution in the sandbox, that's a way for you to do it. And then we are logins. And I think that's for me the most exciting and the greatest feature about the managed agents API because credentials are super hard and super easy to get wrong with agents. And I think the previous talk already also mentioned that we are very careful about using agent because things could go wrong very easily, and we thought a lot about this with managed agents, and we have support for three different ways of credentials, which makes it secure for your agent to use them, and uh provide an isolated way. So, we have bearer tokens for normal API calls, very familiar if you are doing software engineering. We have OAuth for all of the consumer-oriented MCP servers or like tools, and then of course we need environment variables if we want to use certain CLIs or other features. And for managed agents, we are separating them via an external network proxy. So, instead of providing the agent the credentials inside the sandbox, we basically allow you to create credentials externally. They are right-only, so once they are written, there's no way for you to retrieve the the credentials, so you can only override them. So, there's no possibility of leaking, and then you define how and where those credentials should be used. So, for Slack, if you want to connect Slack, you can define, "Hey, here's the credential, and you are only allowed to use that credential with a certain domain or with a certain uh environment variable." So, that way you can control when a credential is used, and that works via this network proxy. And by network proxy, we basically mean a proxy that sits around the sandbox that watches all of the network traffic going in and out. So, if you define credential as an environment variable, the model has a local dummy value, so the environment variable is set, and then when it makes that call, it uses that dummy value, and the network proxy basically uh monitors it, sees, "Hey, here is a uh domain which has to which the user has defined," and then it swaps that dummy value with the real credential, and you can make the call successfully, and then it the agent gets back the result, and has like no idea that this dummy token is not working. And the benefit here is that there's no way for a prompt injection, right? Normally, when you have some malicious attacker, it can tell you like, "Hey, please print all of your values in your .env file." And then the agent might go there and like prints all of your keys, and that's nothing we want to do. So, with this network proxy, where you can define the credentials per domain, it is not possible that an attacker could steal credentials which are getting swapped on your request because the proxy only swaps them for the domains you define. So, there's no way the an MCP server which is malicious can like just steal your Gmail token, which is very cool because it makes it much more easier to trust the agent and to integrate MCP servers which the agent then obviously can use.确定性防护规则（Hooks）与主动触发机制
如果这些还不够，我们还提供了 Hooks（钩子）。Hooks 允许你以确定性的方式（Deterministically）拦截并保护某些决策。如果智能体试图运行某些你希望监控的命令，或者某些需要人工介入审批（Human-in-the-loop Approval）的操作——而这些审批并不属于常规议程对话的一部分——你可以定义 Hooks。它的定义方式与其他 Agent Harness 非常相似：设置一个匹配器（Matcher，本质上是正则表达式），随后智能体生成的每一次工具调用都会检查是否命中规则并触发这些 Hooks。这些 Hooks 同样以文件形式存在，但它们受到受保护权限保护，智能体无法直接访问并修改 Hook 文件来绕过约束。
最后但同样重要的一点是，我们需要一种让智能体变得更加主动的方式。目前我们支持一种触发器类型：基础 Cron 定时触发器。你可以设置定时计划，例如每月 30 天中每天上午 8:00，让智能体在相同的环境中携带特定的输入执行触发任务，以执行尽职调查或审计等工作。
我们目前还在开发 Webhooks 支持，以便实现更高程度的交互性。你可以连接来自特定 SAC 供应商或其他合作伙伴的外部 Webhooks，这些事件将直接触发调用你的智能体；智能体先完成初步审计，然后直接呼叫你的 Slack 或 Microsoft Teams，整个过程完全不需要你亲自管理任何底层基础设施。
这就是整体蓝图。现在让我们来看一些真实的 Demo，并希望现场的网络连接足够稳定。
英文原文
So, and then if that's not enough, we have also hooks available. So, hooks allows you to basically deterministically guard certain decisions. So, if the agent wants to run certain commands which you might want to monitor or where you need some human in the loop approval, which is not part of like the normal agenda conversation, you can define your hooks very similar to all of the other agent harnesses where you have a matcher, which is basically regex, and then every tool call which the agent generates checks if there's a match, and then writes those hooks. And those hooks are also files, but they are protected, so the agent cannot just go there and changes a hook and like basically tries to to to find a way around it. Okay, so last but not least, we need a way to yeah, get our agent more proactive. And currently, we support one kind of triggers, uh basic cron triggers which you can set up a schedule every uh 30 days a month at 8:00 a.m. you want the agent to run a certain trigger with a certain input in the same environment to do maybe some due diligence, some auditing. And we are also working on webhooks that you can even like make this much more interactive that you could connect um external webhooks like from some certain SAC provider or maybe other um providers you work together with which would directly then call your agent. The agent does some first auditing, and then like basically can call your Slack or your Microsoft Teams directly without you even needing to manage some infrastructure at all. So, that's that. Uh final pit picture, and now let's look at some real demos, and let's hopefully hope that the Wi-Fi is good enough.实机演示：自动审计与 Slack 协同交互
我告诉 Gemini 我想做一个很酷的 Google Sheets 演示，它帮我在浏览器中完整复刻了这样一个 Google Sheets 界面。当我点击这里时，实际上就是在模拟前端触发。我们向 Managed Agents 发送真实的 API 调用，指定它使用我们的投资组合审计技能并运行该请求。第一步它看到：“我需要先读取该技能。”
它获取到返回结果，试图理解在这里应该执行什么操作。它依次调用金融数据服务，检查“市场或股票价格是否有变动？”最终，期望它能够完全自动地更新我们自己的 Google Sheets。它仍在处理中，并得出了一些初步结果。
随后它根据该触发事件更新了 Google Sheets。而这个触发器原本可以在每天上午 8:30 自动执行，无需你人工干预。此外，如果智能体拥有 Slack 的访问权限，它甚至可以在特定的交易频道中发布动态：“发生了什么变动？目前的行情如何？其运行机制是什么？”
这里我们还有一个示例，我可以直接向 Spark 询问任何投资建议。希望这也能顺利运行。这需要一点时间。大家在另一个 Demo 中已经见过了……太好了，这里显示有三个持仓仓位。这些分析都是基于 Slack 的上下文生成的。它获取了以往轮次的所有对话记录，给出了发生变动的背景分析，同时依然根据你给出的系统指令克制操作、不越权采取行动。
运行该智能体所需的全部基础设施都包含在 Managed Agents 环境之中。因此，你能非常轻松地上手，将其整合到你已有的工具链中，提供所需的系统配置并开始运行。
回到幻灯片。总结一下数字同事所涵盖的整套服务：
- 我们拥有规划核心：Gemini 作为大脑；
- 我们拥有基于 Managed Agents 的计算机环境；
- 我们可以提供技能配置，复用你在本地电脑上使用的智能体能力；
- 你可以使用工具，通过挂载 MCP 服务连接外部服务与数据提供商；
- 我们提供了安全的凭证管理机制来注入和轮换凭证；
- 我们提供了 Hooks 以确保对确定性逻辑的严格保护；
- 当然，我们还拥有触发器来实现更加主动的系统集成。
非常感谢大家的聆听，希望大家能够尽快去“雇佣”自己的第一位 AI 数字同事。