Legora CTO访谈：从法律检索到智能体编排与法律科技工程实践

人们都在问我们关于 Mythos 以及所有事情，尽管我认为现实中很多法律工作并不需要 [音乐] Mythos 级别的智能。律师助理没了，没错；初级律师没了，我不需要他们了。嗯，那其实并不是我们所看到的情况。我们曾经遇到过一个问题，就是它无法正确引用，所以我们当时觉得，下一个模型发布就能解决。它直接从大概 70% 提升到了大概 99%。就像是我们什么都没做，然后就很棒。但如果你在做一笔比如 4 亿美元的并购交易，你会想要 Legora。我们公司在接下来的历史中就保持 20 名工程师，并且我们要击败所有比我们庞大得多的人。嗯，我们现在有 100 名工程师了，而且感觉我们还远远太小。但我正在琢磨的一件事是，我们的产品中有多少可以纯粹就是那个智能体，或者完全是其他那些东西。

英文原文

People were asking us about mythos and everything, even though I think realistically a lot of legal work does not require [music] mythos level intelligence. Paralegals gone, yep. Junior lawyers gone, I don't need them. Um, that's actually not what we're seeing. We had one problem where like it wouldn't cite properly and so we were like, next model release solved. It went from like 70% to like 99. It's like we did nothing and like, great. But if you're doing a, you know, $400 million M&A transaction, you're going to want Legora. We're just going to be 20 engineers for the rest of the company's history and like we're going to beat everyone who's much bigger. Um, we have 100 engineers now and it feels like we're way too small. But it's something that I'm toying with is like how much of our product can just be the agent or solely the other stuff.

[音乐]

英文原文

>> [music]

对话开场与团队背景

好的，我们现在在演播室，和特邀嘉宾——Legora 的首席技术官 Jacob Loritzen 在一起。我说对了吗，Loritzen？还行，是的。没关系，我对怎么读不是很讲究，它是丹麦语。对，没错。它有什么特殊的渊源吗？

英文原文

>> Okay, we're here in the studio with special guests, uh, Jacob Loritzen, CTO of Legora. Did I get it right, Loritzen? That's fine, yeah. That's okay. I'm not too particular on how it's Danish. Yeah, exactly. Does it have a special history?

它就——不，sen 就像是儿子，“谁的儿子”。所以就像英文里的 son 一样，是的。所以就是 Loritz 的儿子，这就是我的名字。

英文原文

>> It just no, sen is like son, son of. So just like you have son and yeah. So son of Loritz, that's my name.

很好。额，我们之前在这里那里有过交集。你——你——你在——额，是的，你发表演讲过几次，显然 Legora 是历史上增长最快的法律科技公司之一。是的。而且，额，你在开场前、我们预备节目时说了些我认为非常有意思的话。你说你们的渊源非常欧洲，额，尽管显然你们正在纽约和旧金山扩张。你们的文化有什么感觉不一样的地方？

英文原文

>> Nice. Uh, we've crossed paths here and there. You you you spoke at uh, yeah, you a few times and obviously Legora is one of the fastest growing legal tech in history. Yep. And uh, you said something at the start in our pre-show that I thought was really interesting. You said that your origin is very European, uh, even though obviously you're expanding in New York and SF. What feels different about your culture?

我认为我们骨子里可能对我们所做的事情非常谦逊。因为从最最一开始，我们一直就是处于弱势地位的一方。没人相信斯德哥尔摩一间地下室里的一家小型初创公司能够，你知道，击败在美国遥遥领先于我们的一尊庞然大物。所以我觉得我们一直都是弱者，因此我们一直都非常谦卑。嗯，我们一直都不得不为了我们的生存而奋斗。嗯，我认为有一点是我们必须去学习的，随着我们也变得越来越，嗯，随着我们进入美国并在美国呆得越来越多，那就是变得稍微更大胆一些，就像是直接向全世界展示我们有多棒。因为我觉得这对于我们中的有些人来说并不自然。

英文原文

>> I think we are just at the core probably very uh, humble about what we're doing. Because we've always been the underdog from the very very beginning. No one believed a small startup in a basement in the Stockholm was going to, you know, outcompete the huge behemoth that was like way ahead of us in the US. So I think we've always been underdogs since we've always been very humble. Um, we've always sort of had to fight for our uh, existence. Um, I think it's something we've had to learn as we've also become more and more um, as we've entered the US and been in the US a lot more is to be a little bit more uh, bold and like just tell the world how great we are. Cuz it's it doesn't come naturally to I think some of us.

产品演进与核心功能

我有点想先从梳理你们的产品开始，因为我觉得对于很多这类法律产品，你知道，它们有点像处于保密状态。比如，你知道，不应该拿出来展示。但是，嗯，是的，如果你能谈一谈产品长什么样？也许还有你们是怎么走到这一步的？比如，到底经历了哪些产品迭代才让你们走到今天的位置？[清嗓子] 我觉得那会非常有趣。

英文原文

>> I kind of want to start by going over you guys's product cuz I feel like with a lot of these like legal products, you know, they're kind of like under wraps. Like, you know, you're not supposed to like show it. But, um yeah, if you could just talk about like what does the product look like? And maybe like how did you guys get there? Like like what have been the product iterations that got you to where you [clears throat] are today? I think that would be really interesting.

好，当然可以。一切开始的方式就是内置了瑞典法律搜索工具的 ChatGPT。

英文原文

>> Yeah, absolutely. The way everything started was ChatGPT with uh Swedish laws

好的。

英文原文

>> Okay.

搜索工具在里面。那简直就是当时的全部。所以，那就是我们起步的地方。我们从法律检索开始做起，这基本上在世界上是一个巨大的市场，嗯，律师们花很多时间去弄清楚之前有没有类似这样的案例。法律先例是什么？比如，这是，你知道，之前的人们在这些情况下做过什么？那极其重要。嗯，所以，那一直是我们产品的一部分。基本上，我们最开始就是一个你可以向它提问的智能体，它会去查询法律。那就是非常简单的开端。嗯，然后从那之后它就拓展了。它拓展到了这个智能体能做什么：它现在会，你知道，非常出色地起草很多东西。它会编辑 Word 文件。它会，你知道，把你的文件归档、整理好一切。嗯，运行非常复杂的分析。嗯，而且我们给它添加了一大堆功能。所以，我们的功能之一叫做表格审查（tabular review），这个名称现在正在被行业所使用。我们非常高兴看到 Casetext for legal 在他们正在构建的东西上使用了我们的术语。

英文原文

>> search tool in it. That was literally the entire everything. And so, that's where we started. We started with legal research, which is like a huge market basically in the world where um lawyers spend a lot of time figuring out what are previous cases that look like this. What's the legal precedent? Like, is this, you know, what have previous people done in these situations? That's extremely important. Um And so, that has always been part of our product. Basically, we started from like an an agent you could ask questions and it would query laws. That was the very simple beginning. Um and then from that it's expanded. And so, it's expanded in like what can the agent do? It it will now, you know, draft a lot of stuff really well. It will edit word files. It will, you know, move your files away, organize everything. Um run really complex analyses. Um And we've added just a bunch more features to it. So, one of our features is called tabular review, which is um now a name that's being used in industry. And we're very happy to see Casetext for legal using our terminology for what they're building.

那个。我们得聊聊那个。我们得聊——

英文原文

>> that. We're going to have to talk about that. We're going to have to talk

[笑声与惊叹声]

英文原文

>> [laughter and gasps]

它本质上就是一种能够批量提取和批量审查很多、很多、很多协议的方式，这是你在尽职调查以及融资或类似场景中要做的事情。

所以，那是它的一部分。就好像我们帮助律师把他们的工作做得高效得多，做得好得多。

然后在单纯做工作之外，我们现在在协助编排工作方面也投入了更多，比起甚至仅仅一年前多得多。所以，我们有不同种类的——我们有一个叫做“列表（lists）”的东西，对我们来说有点像 Linear。嗯，我们有“录入（intake）”，这是一种与客户协作的方式。我们还有“门户（portal）”，这也是一种协作的方式，你知道，律师对律师、客户对律师、律师对智能体、或许还有智能体对智能体。我们在这上面花了很多时间，因为坦率地说，你 10 年前或者甚至今天与律师合作的方式真的很痛苦，说实话，在这方面有巨大的空间可以用好得多的方式去做。所以这些我认为是两件事：一件是做具体的工作，另一件是编排这些工作。

英文原文

>> Which is essentially like a way to just bulk extract and bulk review many many many um agreements, which you do in due diligence and when you raise money or whatever.

So, that's part of it. It's like we help lawyers just do their work much more efficiently, much much better.

And then outside of just doing the work, we're also now um helping orchestrate the work a lot more than we were just even a year ago. So, we have different sort of we have a a thing called lists, which is sort of like linear for us. And um we have intake, which is a way to collaborate with um clients. And we have um portal, which is also sort of a way to collaborate, you know, lawyer to lawyer, client to lawyer, lawyer to agent, agent to agent, maybe. We're spending a lot of time there cuz um there's just frankly the way that you work with a lawyer 10 years ago or maybe even today is is really painful and there's loads of room to do that in a much better way to be honest. And so those are like I think the two things is it's doing the work and it's also like orchestrating the work.

从助手到智能体与底层模型

你们是什么时候开始做或者构建智能体的？

英文原文

>> When did you guys start doing or building the agent?

所以我的意思是，我们基本上经历了从我们所谓的“助手”到“智能体”的转变，对吧？嗯，我猜智能体大概，我不知道，我想大概是一年前吧。

英文原文

>> So I mean we've all we basically had the switch from the assistant to the agent we call it, right? Um I guess that the agent I don't know I want to say a year ago probably.

是的。

英文原文

>> Yeah.

所以给想要理清时间线的人提供一个大致的脉络：YC 2023 年——

英文原文

>> So this is a rough timeline for people who are understanding YC 2023

是的。

英文原文

>> Yes.

然后你也是 YC 的创始人。

英文原文

>> And then you were you were a YC founder as well.

YC 24。

英文原文

>> YC 24.

YC 24。

英文原文

>> YC 24.

是的，我是 YC 23。对，Legora 是 24。

英文原文

>> Yeah, I was in YC 23. Yeah, Goro was 24.

好的。

英文原文

>> Okay.

所以这有点在智能体真正爆发之前。

英文原文

>> And then so so there is this is like a little bit pre pre like agents really taking off.

确实。

英文原文

>> Exactly.

然后使用模型显然改变了一切。

英文原文

>> And then like using models obviously change everything.

是的。

英文原文

>> Yes.

我的意思是，你知道，从某种意义上来说，今天，你知道，我们看过了产品预览。它确实看起来——好吧，预览，我们看了一个私下的展示，但我们会把视频放到 YouTube 上。

它基本上很像面向律师的 ChatGPT，对吧？这样说非常公允，但侧边栏有许多非常棒的定制工具，比如表格视图，比如我非常喜欢的录入，然后你知道，还有一堆其他东西。

英文原文

>> I mean you know in one in one sense today you know we had a preview of the products. It does look well preview. We had a private thing but we'll put up a video on on YouTube.

It's mostly like chat GPT for lawyers, right? Like a very like very fair to say that but then like the sidebar has a lot of nice custom tools like the tabular view like the in intake which I really liked and then you know a bunch of other things.

没错。那——那完全就是它的本质。它是——它是同事般的 ChatGPT。

英文原文

>> Exactly. That's that's exactly what it is. It's it's co-worker chat GPT.

它对于法律工作来说真的非常、非常出色。

英文原文

>> It's like a really really good for legal work.

然后是的，针对比如企业法务、寻找即将出台的法规的人们等等的工作流程，我们可以多谈谈那里面的产品。

英文原文

>> And then yeah the the the workflow for like either like in-house corporate lawyers like people like looking for upcoming regulations all these things and we can talk about more of the the products there.

对了，你们底层在使用什么模型？能谈谈这个吗？

英文原文

>> Yeah, what models are you guys using underneath? Like can you talk about that?

我的意思是，我们根据任务的不同使用一堆不同的模型，对吧？那是，嗯——

英文原文

>> I mean we're using a a bunch of different ones depending on task right? That's um

然后那是律师看不到的。你们自己来挑。

英文原文

>> And then that's not available to the lawyer to see. You guys pick yourselves.

是的，嗯，我们在这个问题上反反复复考虑过，是的，我觉得就像其他人一样。在智能体里，主模型是有一个模型选择器的，但在某些工具中我们会使用不同的模型，还有比如表格视图；而且像——我们尝试过，我认为我们发现最有效的方法是细分问题，并在我们细分任务时进行路由，但我们目前在智能体中确实有一个选择器。

英文原文

>> yeah, well we've been back and forth on this yeah just like everyone else I think. In the agent there's a model selector for the primary model but there are certain tools that we use different models in and there's like tabular view and like we try I think we found it most effective to subdivide the problem and and and route on like as we subdivide the task but we do right now have a a picker in the agent.

嗯。

英文原文

>> Mhm.

说实话，我不确定它是否会一直留在那里。嗯，所以是的。

英文原文

>> I'm not sure if that's going to stay there always to be honest. Um so yeah.

是的，我的意思是，我对这个带来一点背景认知，因为我是 Cognition 的顾问，当我加入时，你知道，关于智能体实验室是什么，我的论点的一部分是：一个智能体实验室，你知道，理想情况下是一个特定领域的 GPT 封装套壳，对吧？就像我们——我们——我们运行——我们是领域专家，我们会关注你，我们会比前沿实验室更加全方位地服务你的需求，而且我们会根据我们的判断为你提供前沿实验室最好的东西。所以我们的职责是替你做决定，因为你不是——你的职责不是——

英文原文

>> Yeah, I mean so I bring a little bit of context to this because I'm advisor for cognition and when I joined you know part of my my thesis for what an agent lab is an agent lab is you know ideally a domain specific GPT wrapper right like we we we run the we are the domain experts we'll pay attention to you we'll serve your needs way way more than the frontier labs and we'll give you the best of the frontier labs according to our judgment. So our job is to decide for you because you're not your job is not to

没错。我们会始终让——

英文原文

>> Exactly. We'll always keep

让你保持在帕累托前沿上，并确保它一直都很出色，而且你不必钻研所有细节，因为我——正是针对这一点，如果去找一个律师，问他们：噢，你要根据这个先例去起草这份合同，你认为——

英文原文

>> you on the Pareto frontier and make sure that it just is great all the time and you don't have to dive into all the details cuz I one of the things that I've exactly to this point if you go to a lawyer and you ask them oh you're going to you know draft this contract based on this precedent do you think

模型选择与法律场景需求

那是一开始就指定更高阶的模型，还是在任务中途切换到 Opus？其实你知道……

英文原文

that's a sign it higher or an opus mid task? Actually you know

［笑］

英文原文

>> [laughter]

他们、他们根本分不清。直接给我用这部分模型就行。只要确保它能跑通就行。我的时薪可是 1,000 美元。总之只要能搞定就行。

英文原文

>> they they won't know. Just give me this part of the model. Exactly just like make sure it works. My hourly rate is $1,000 an hour. Exactly just like make it work.

［笑］

英文原文

>> [laughter]

嗯，所以是的，我最初进来的时候也是那么想的，当时我觉得“噢对，这就是 Devin 的不同之处”。而现在我们的模型选择器里有五种选项，包括 Fable、Opus 和 Astra，人们就是想拿时髦的新模型来试试。

英文原文

>> Um so yes I originally came in with that I was like oh yeah this is this is what Devin does differently and now we have five things in the model picker including Fable and Opus and Astra and people want to experiment with the fancy new model.

对。

英文原文

>> Yeah.

你们那边也是这样吗？

英文原文

>> Do you is that the same?

我觉得是的。我想这在法律领域大概、肯定要比在编程工程领域少得多。

但确实有些人一上来就只想要 Fable 1；还有人会问我们关于 Mythos 以及各种模型的事情，嗯，尽管我觉得现实中很多法律工作其实根本不需要 Mythos 级别的智能。

英文原文

>> I think so. I think it's probably it's it's definitely way less in legal than it is in coding engineering.

But there are definitely people that would just want Fable one right away and you know people were asking us about Mythos and everything um even though I think realistically a lot of legal work does not require Mythos level intelligence.

没错。就是这样，对吧，我觉得基本就像是：“行吧，我们给你这个选项，这是你要的，但你消耗点数的速度会快上差不多 10 倍……”

英文原文

>> Yeah. There is so exactly right like I think basically you're like okay we'll give you the option you you asked for this but you're going to run down your credits like 10x faster than

是的。我本来很希望……我们抗拒引入这个选择抗拒了很长时间，最后我们还是有点妥协了，觉得“好吧，如果你想要这种掌控度，你知道的，那我们就提供给你”。

英文原文

>> Yeah. I would have loved we we fought introducing it for a very long time and then in the end we we just kind of caved in and we're like okay if you want this level of control you know then we'll give it to you.

对。那你觉得现在的模型是不是已经完全满足律师的需求了，比如根本不需要再变得更好了？

英文原文

>> Yeah. So you think the models are already like as good as lawyers need like there's no need for them to get better?

这是一种……我认为是一种幂律分布。可能大约 60% 的工作我们现在就已经能做得非常好了，从现在起，更多是成本和延迟方面的问题。

但剩下的那部分，特别是非常复杂的分析，我们确实仍然需要更高的智能。而且我觉得工作本身也会扩展，好比说，这就是我们目前手头的工作，但业务也会顺着律所展开，因此大概率会随着智能水平的提升而不断扩充。

英文原文

>> It's a I think it's a power distribution so like there's probably like 60% of the work we can do really well right now and the you know from now on it's like a cost and latency thing.

But then there's the rest of the the thing where especially for like very complex analysis, we we still need more intelligence. And then I think there's also just the work expands kind of like that's the work that we have right now, but also the work will follow the law firm, thus probably expands with intelligence.

确实。

英文原文

>> Yeah.

我想是这样的。

英文原文

>> I imagine.

评测体系与法律工程师

早期的时候，比如你们刚起步、我猜是向智能体转型或者构建智能体的时候，你们当时都在用什么模型？我猜大概是 Opus 46 之类的？

英文原文

>> Early on like when you guys first started, I guess transitioning to the agents or building the agents, like what were like what models were you guys using? I guess it would have been like Opus 46 or something.

那会儿应该对，Sonnet 46、Opus 46 差不多是对的。

英文原文

>> That would have been yeah, Sonnet 46 Opus 46 is probably right.

当时智能体不擅长什么，而现在即使仅仅过去了 9 个月，它们就已经做得好得多了？

英文原文

>> What were the agents like not good at at that point that they're much better at now even though it's only like 9 months later?

子智能体委托就是其中之一，我们现在把它做得非常成熟了，特别适用于展开式的检索查询，目前对我们来说效果极好。嗯，这在以前是很困难的。我们基本上一直在做追踪，所以我们有一个专门负责评测（evals）的团队，我们会追踪所有模型在不同执业领域的表现——也就是法律的各个细分方向——以及贯穿所有这些领域的不同任务。所以我们基本上有一个很大的热力图——

英文原文

>> Sub-agent delegation was one of them that we've gotten working really well now for like fan out type retrieval queries, which works extremely well for us now. Um that was difficult before. We basically track so we have a team that's just doing evals and so we track all the models' performance in different practice areas, which is like subgroups of of legal and different tasks across all those. So we basically have like a big heat map

比如我们在上面……

英文原文

>> where we're like

嗯哼。

英文原文

>> Mhm.

然后我们就能看到它们的变化。通常情况下它们在绝大多数项目上都有进步，但其实有相当多的模型版本发布时，是在某些方面进步了，而在其他方面退步了。

英文原文

>> And then we can see them move and it's like usually they improve most stuff, but there's actually a lot of model releases where they improve some and and regress in others.

好的。

英文原文

>> Okay.

而且这种情况发生的频率远比你想象的要多。

英文原文

>> And that actually happens a lot more than you'd think.

比如具体有哪些任务呢？你对你们跑大量评测的具体方式熟悉吗？就是说，你们观察的那些不同的细分任务到底是什么样的？

英文原文

>> Like what are some of those tasks? Like are you familiar with how you guys are running a lot of the evals? Like what like what are those different sub-segments I guess that you look at?

比如会有……会有各种不同维度的设定。我们基本上有不同的“世界”，本质上就像是我们的环境。因为律师或多或少总是在围绕项目工作，所以就像是：这里是一个有一万份文件的数据室，你正在做尽职调查；最终一份优秀的报告长这样——它包含了所有这些预警信号（red flags），没有包含任何奇怪的内容，而且我们希望它言简意赅，等等。我们差不多就是这样搭建评测的。所以这里有不同的任务，其中一部分是起草报告：草稿看起来好不好？是否包含了所有正确的内容？措辞是否得当？分析是否合理？还有一部分是发现预警信号，也就是起草之前的分析。所以我们差不多可以对它进行细分。我们就是这么跑评测的。

英文原文

>> So like there's um there's different just like framings of So we have different basically like worlds like our environments essentially, which is like here's a cuz cuz lawyers work on projects always more or less. So it's like here's 10,000 file data room. You're doing a due diligence. In the end a good report looks like this. Like it contains all these red red flags and like it doesn't contain whatever weird stuff and like we want it to be concise, etc., etc. That's kind of how we've built it. So there's different tasks, which is like part of it is drafting a report. Like does the draft look good? Does it contain all the right stuff? Is the wording sound? Is the analysis sound? There's also the part of finding the red flags, like the analysis before that. So we can kind of like subdivide it a little bit. And that and that's how we run the evals.

很有意思。等等，你们的评测团队是什么架构？你们是雇律师来做这个吗？

英文原文

>> Interesting. Wait, what does your evals team look like? Like do you hire lawyers to like do that?

是的，人员是混合的。我们有工程师，也有律师，他们组成同一个团队。目前大约有三名工程师和三名律师；然后我们还有一个人才库，里面是我们称为“法律工程师”的人员，他们是在大型律所（big law）工作过大概三四年的人。通常他们可能拥有计算机科学学位，在技术上非常前沿，他们来到我们这里是因为希望协助改变这个行业。

英文原文

>> So yes, it's a mix. We have uh engineers and then we have lawyers and it's one team. So there's like three engineers and like three lawyers right now and then we have a pool of what we call legal engineers, which is um people that have worked in big law for like three or four years. Typically maybe they have a CS degree and like they're very tech forward and then they come to us and because they want to like help change uh the industry.

他们在大型律所里是做 IT 的吗？还是大型律所里的执业律师？

英文原文

>> Are they IT in big law? Or are they lawyers in big law?

他们是大型律所里的律师。

英文原文

>> They're lawyers in big law.

带着计算机科学学位。

英文原文

>> With CS degrees.

没错。天啊，这个韦恩图重叠的部分……是的，完全没错。这简直是最难招的一类人了。

但他们也是极其强大的力量倍增器，因为我们可以调动他们来做评测工作和产品研发。这就像是你拥有一个由上百位专家组成的人才池，他们每天使用这个产品，每天去和客户沟通。而且他们亲身过着这种生活，也经历过这些。

我们随时可以调动他们来做评测，帮我们打磨产品，这是一个巨大的优势。

英文原文

>> Yes. Oh my god. It's a Venn diagram that's Yeah, exactly. That's like the hardest person to hire for.

But they're also such a force multiplier cuz we can pull on them for the eval stuff and for the product development. It's just like you have a pool of a hundred experts that use the product every day and go and talk to clients every day. And also like they live this and have lived this.

And we just pull on them to do evals and just like help us shape the product and it's such a such an advantage.

工具打磨与法律行业的未来范式

很想了解一下，你们是什么时候建立那个团队的？

英文原文

>> And curious like when did you guys start that team?

评测团队是在某个阶段从助理智能体团队里脱胎出来的。我想直到不久前都存在一个阶段，坦白说在那时做评测其实不太值得，因为所有的模型本身都在变得越来越好。所以当时的问题是：

我们是想花大量时间去做非常复杂的评测，还是直接去构建产品？因为单单通过打造一套更好的测试与运行脚手架（harness），就能获得巨大的提升。而且只要顺应“一切都会变好”的趋势，一切就真的自然而然地变好了。

英文原文

>> Evals um grew out of the assistant agent team at some point. I think there was a point up until recently where it was like honestly just not really worth it cuz like all models just got better. So it was like

do we want to spend a lot of our time doing really complex evals or do we just like want to build cuz there was so much to be gained from building a better harness. And like just sort of leaning into everything is going to improve and everything kind of just improved.

嗯哼。

英文原文

>> Mhm.

是啊，就像每隔两三个月你就能拿到一个新版本的模型。

英文原文

>> Yeah, like every two three months you get a new version of the model.

对，完全是这样。

英文原文

>> Yeah, exactly.

直接接上去用就行了，对吧？

英文原文

>> Just pass it through, right?

没错。我是说我们之前遇到过一个问题，模型没办法正确给出引用，导致我们很难顺着线索追踪；结果下一个模型版本发布就解决了。准确率直接从大约 70% 飙升到了 99% 左右。这简直就像我们什么都没做，效果就很棒了。

英文原文

>> Exactly. I mean we had one problem where like it wouldn't cite properly and so we would difficult to follow and it was like next model release solved. It went from like 70% to like 99. It's just like we did nothing and like great.

是啊，也许真正的核心工作在于保证所有数据的私密性，让数据可打标签、可排序，添加像你们做的那种表格视图之类的新功能。这就是……也许……

英文原文

>> Yeah, maybe the the the real work is uh you know, keeping all the data private, making it taggable, sortable, adding new features like the the tabular view that you made. Is that the you know, maybe

你的意思是在 ChatGPT 之上额外做的那些工作？

英文原文

>> You mean over and above ChatGPT?

可以这么说，但显然现在你们也开始做评测了。我猜你们对开源模型的使用也在增加，这个我们之后可以聊聊。所以是的，我也不知道我除了这些还想表达什么，比如：这份工作的本质是什么？这份工作……

也许在早期并不需要做太多的评测，因为当时主要是把模型的能力直接开放给律师。

英文原文

>> Like yes, but obviously now you're also doing evals. I assume your your use of open models is also increasing. We can talk about that. Uh so yeah, I I don't know what I'm saying here apart from like you know, what is the job? The job is uh

maybe not uh in the early days it was not so much to do do evals because it was mostly exposed model capability to lawyers.

是的。其中很大一部分工作其实就是弄清楚在有了这一切之后，法律行业在 5 年后、10 年后会是什么样子？比如编程领域可能领先了一到两年，所以我们可以稍微参考一下它；但核心的基本原语是什么？人们怎么工作？如果你不再是用 200 封邮件的往来线索、带着 V5 版本的 Word 文件改来改去，那应该怎么协作？那会是什么样子？这甚至不一定纯粹是 AI 的问题，而是在于构建出优秀得多的产品体验。

英文原文

>> Yeah. The huge part of it is just um, figuring out what does the legal profession look like in 5 years, 10 years with all of this stuff? Like coding is maybe a year or two ahead so that we can kind of look to that a little bit but like what are the primitives? How do people work? How do you collaborate if it's not a 200 thread email with like V5 word files back and forth? Like what does that look like? And that's like not even necessarily purely AI. That's just like build a much better product experience.

对。

英文原文

>> Yeah.

并且确保你在做这些的同时具备审计日志、数据留存，以及所有在这类受监管行业中非常关键的事项。

再然后才是 AI 的部分，也就是让工作变得更快更好，以及打造能够在新事务进入时主动做出反应的主动型智能体，这样你就不必……你知道的，不必事事都去人工启动。所以我觉得很多工作单纯是在思考：我们认为 5 年后它会变成什么样？如果模型能力持续增强，如果价格持续下降，我们能做什么？你登录进去之后，它会不会像一个指挥中心，直接呈现出“智能体最近一直在处理这五件事，你需要做出响应”，然后由你来审核、检查并改进这个系统。也许这就是我们的方向，或者正在迈向的终点。

英文原文

>> And make sure you do it with like auto logs and data retention and all of those sort of really important things in regulated industry.

And then there's the the AI part which is like make the work faster and better and and proactive agents that react to stuff when it comes in so you don't have to, you know,

kick off everything. So I think a lot of the work is just figuring out what do we think it's going to look like in 5 years? If, you know, model capability keeps increasing, if price keeps coming down, what could we do? Could you log in and it's like command center where it's just like you need to react to these five things that the agents have been doing lately and like review and check and improve the system. Maybe that's where we're going or is where we're going.

是的，没错。对，我刚才正好想问：你能稍微深入讲讲那方面吗？

英文原文

>> Yeah, yeah. Yeah, I was going to ask like could you go into that in a little more depth?

比如，对，几年后法律行业会是什么样子？

英文原文

Like yeah, what what is the like law profession going to look like in a few years?

[笑声]

英文原文

>> [laughter]

我得先说明我不知道。我是说，谁真的知道呢？不过，嗯，我觉得我们可以多参考软件编程。我认为，嗯，所有律师的效率都会比以前高得多。他们将能以更低的成本完成更多、质量更高的事情。我认为这个行业本身、工作本身会得到极大改善。会变得更加——我认为律师会把更多时间花在他们当初成为律师想做的事情上。呃，你知道，搞清楚什么是合适的风险状况、哪些是可控的风险、什么是——发挥他们全部的判断力、战略判断，以及比如与客户沟通。他们会更多地做这些事情，而更少地去做那些比如在浩繁的文件中搜索确凿证据，或者像很多律师一直在做的那种繁琐、超级烦人的工作。

英文原文

>> I should preface this by saying I don't know. I mean, who knows really? But um, I think we can look to coding a lot. I think um, all lawyers are going to be way more productive than they were before. They're going to be able to do a lot more stuff at a higher quality for lower cost. I think the profession itself, the work itself is going to improve a lot. It's going to be a lot more I think lawyers are going to spend more of their time doing the stuff that they became lawyers to do. Uh, you know, figuring out what's the right risk profile, which manageable risk, what's like the all of their judgment, strategic judgment and like working with clients. They're going to do that a lot more and much less of the like trolling through big documents like find the smoking gun or like do all this menial super annoying work that a lot of lawyers have been doing.

合伙人模式的演变

你觉得合伙人模式会发生什么变化？比如你觉得它会保留下来，还是会怎样？

英文原文

>> What do you think's going to happen in the partnership model? Like do you think that stays or like what happens?

我对这个——我其实上周刚对这个做过很多思考。就是合伙人模式——我不想深入太多细节，但合伙人模式之所以存在，是因为这些公司拥有的唯一资本类型就是人力资本。所以这并不像，你知道我们，或者像软件工程公司，或者像制造业不管什么，你必须有物理资产，或者某种随着时间推移积累起来的无形资产，它们能为股东带来分红。但如果你在专业服务这样的行业工作，像咨询公司等等也是一样，你所有的价值都在于你的人，而你无法真正拥有人力资本。你知道，你你不能——你知道，奴隶制已经被废除了，你不能拥有人，但你可以租用他们，你可以付钱给他们。因此，合伙制本质上是一种让这种人力资本成为公司所有者的架构，好让他们留下来。

英文原文

>> I gave this a I actually gave this a lot of thought just like last week. Just the the partnership model I don't want to get into too much of the details but like the partnership model kind of exists because the only type of capital that these companies have is human capital. So like it's not like you know us or like a software engineering firm or like a manufacturing whatever you have you have to physical assets or like some sort of intangible assets that that you accrue over time and and they pay dividends to the shareholders. But if you work in an industry like professional services and it's the same for like uh consultancies and and whatever else, all of your value is in your people and you can't really own human capital. You know, you you can't you know, slavery's abolished, you can't own people, but you can rent them, you can pay them money. And so partnerships essentially is a structure where you make the human capital the owners of the company so that they stay.

是的。

英文原文

>> Yeah.

这种情况会改变吗？比如我不——它不一定会变，但我觉得我们真正想和律师一起做的是，我们实际上希望律师和律所能够建立某种无形资本、某种持久的资本。比如你可以构建一个学习你工作方式的系统。现在很多律所会认为，存在一种“某律所的工作方式”。我认为我们可以将其规范化编码。我们可以构建一个系统，无论它是一个自我完善的知识图谱还是什么，以及流程、文件和档案，那里有一种能够产出非常出色成果的工作方式。那就可以成为他们的知识产权。所以这就像是一种将这种人类专业知识和判断力编码进系统的方法。我认为那是——现在我完全是在漫无边际地说了，但比如，嗯，那是我们想要帮助赋能的东西。

英文原文

>> Could that change? Like I don't It doesn't have to, but I think what we really want to do with the lawyers is we actually want lawyers to be able and law firms to be able to build some sort of intangible capital, some durable capital. And like you can build a system that learns of how you work. A lot of firms right now think about like well there's the X law firm way of doing things. And we can codify that, I think. And we can build a system and whether that's like a self-improving knowledge graph or whatever it is and like process and files and documents where there's like a way of working that produces a really good out output. And that can be their IP. And so it's like a way to codify this human expertise and judgment into a system. That I think is Now I'm just rambling completely, but like um that's something that we want to help empower.

嗯。

英文原文

>> Mhm.

那这会改变合伙人模式吗？因为这样一来激励机制就不同了。在合伙人模式中，你只需要拉来更多客户，你不想为了将你的信息编码化而耽误招揽客户的时间，基本上是这样。

英文原文

>> And does that change the partnership model then cuz then incentives are different. In the partnership model you just got to bring more clients and you don't want to take away time from bringing in clients to codify your information, basically.

嗯。

英文原文

>> Mhm.

因为像你的回报来自于，你知道，“多劳多得”（eat what you kill）或者搞这种资历逐级晋升（lockstep）之类的。嗯，这需要做些改变。我们需要找出一种激励结构，让合伙企业——呃，合伙人受到激励去对事物进行系统沉淀，为公司构建资产。但目前并没有，因为一旦他们离开合伙企业，他们就，嗯，拿不到他们所构建的资产的任何分红了。

英文原文

>> Cuz like you get rewarded from you know, eat what you kill or make this lockstep whatever. Um that needs to change a little. We need to figure out an incentive structure where partnerships uh, partners are incentivized to codify things for to build an asset for the company. But right now there aren't cuz the sooner as they leave the partnership they don't, um, get any dividends from the assets that they built.

是的。

英文原文

>> Yeah.

所以这是个，是的。我觉得整个商业模式是，呃，我们走着瞧吧。

英文原文

>> So that's like a, yeah. I think the the the whole business model is is, uh, we'll we'll see.

初级岗位与金字塔结构

是啊，是啊。[笑声] 现在一切都成了疑问。是的。我实际上另一个看法，你知道，在合伙人层面，那是行业的顶端。其实我的看法更多是在底层，对吧？

英文原文

>> Yeah, yeah. [laughter] It's all It's all in question right now. Yeah. My My actual other take, you know, so on the partnership level that's like the top end of the the industry. Actually, my take is more on the bottom end, right?

底层怎么了？

英文原文

>> What happened there?

呃，律师助理没了。

英文原文

>> Uh, paralegals gone.

是的。

英文原文

>> Yeah.

对吧？比如初级律师没了。我不需要他们了。

英文原文

>> Right? Like junior lawyers gone. I don't need them.

是的。嗯，实际上我们看到的并不是这样。所以对，它是——我觉得它是，我不知道。你会说我们在工程领域看到这种情况了吗？

英文原文

>> Yeah. Um, that's actually not what we're seeing. So yeah, it is I think it's I don't know. Would you Would you say that we're seeing that in engineering?

呃，是的。初级工程师真的可以说是完蛋了。就像你必须从大学出来或者辍学时就是完全成熟的，否则你——

英文原文

>> Uh, yeah. So junior engineers are really kind of cooked. Like like you kind of have to come out of uh, drop out of university fully baked or else you're

[笑声]

英文原文

>> [laughter]

否则你根本混不下去。嗯，我觉得我有点为了喜剧效果而夸张了。但像是，呃，对，我认识很多律师，他们从律师助理做起，去读法学院，经过这一切，对吧？而现在他们不——不需要他们了。他们真的就是不需要了。

英文原文

>> or else you're not going to make it. Um, I feel like I'm I'm being a little dramatic for for uh, comedic purpose. But like uh, yeah, I know a lot of lawyers they started as paralegal, they go to law school, all these things, right? And now they don't they're not needed. They're they're seriously just not needed.

是啊。是啊。对，合伙人模式以及金字塔结构会发生什么，这很有意思。我认为大体上很多律师，嗯，你仍然需要你的——你需要下一代。至少在工程领域，我的看法是，你现在拥有一个优秀的工程师，他们现在的价值比两年前高得多，因为他们能产出多得多的东西。如果他们很优秀的话。我认为在法律领域也是同样的情况，如果他们很优秀，他们实际上可以做很多非常出色的工作。所以我认为更确切地说，真正优秀的律师或初级律师身价高得多了。

英文原文

>> Yeah. Yeah. Yeah, it is it's interesting what happens to the partnership model and like the pyramid. I think generally a lot of lawyers, um, you still need your you need the next generation and at least for engineering the way I think about it is you have a good engineer today, they're worth a lot more even than they were 2 years ago cuz they can produce so much more stuff. Like if they're good. And I think it's the same thing in legal where like they can actually do a lot of really good work if they're good. So I think it's more that like really good lawyers or junior lawyers are worth a lot more.

是啊，一直以来都是这样，对吧？是的。

英文原文

>> Yeah, it's It's always been the case, right? Yeah.

是的，但现在更是如此。就像——是的，两极分化更加严重了。

英文原文

>> Yeah, but it's even more so. It's like it's Yeah, it's polarizing even more so.

计费模式与高价值工作

我其实对此很好奇。整个法律业务基本上就像是，嗯，按小时收费。

英文原文

>> I'm actually curious about that. So like the whole like law business is basically like, um, charging per hour.

嗯。

英文原文

>> Mhm.

所以——

英文原文

>> So

是的。

英文原文

>> Yeah.

你知道，如果每个项目现在都能快得多地完成，它会变得便宜得多吗？比如——

英文原文

>> you know, if each project now can be done a lot faster, is it going to be like way cheaper? Like

这实际上，所以——

英文原文

>> This was actually so,

嗯，

英文原文

>> Um,

有趣的故事。我当时在读 Max 的 YC 申请，帮他进入 YC，让当时名为 Leia 的项目进入 YC。这也是我当时提出的一个疑问——

英文原文

>> fun story. I was I was reading Max's YC application and like helping him getting into YC get Leia as it was called back then it's YC. And this is why one thing I was like

好奇问下，那是丹麦语里的什么词吗？

英文原文

>> And curiously is a Danish thing for something?

不，是瑞典语。

英文原文

>> No, it's Swedish.

瑞典语？

英文原文

>> Swedish?

不，其实它什么意思也没有。有一个非常搞笑的故事，我记得是一家西班牙报纸在早期报道 Leia 时，他们解释说这个缩写代表“合法高效的尤达助手”（legally efficient Yoda assistant）。

英文原文

>> No, also it means nothing. There's this really funny story in I think it was a Spanish newspaper that wrote about us in the early days Leia and then they they explained that the abbreviation was legally efficient Yoda assistant.

[笑声] 为什么是尤达？

英文原文

>> [laughter] Why Yoda?

是啊，我也不知道。我也纳闷为什么是——

英文原文

>> Yeah, I don't know. I was like why why is it

看吧，幻觉不仅来自洛杉矶（LA）。

英文原文

>> See the hallucinations are not just from LA.

不，我认为是记者产生了幻觉。

英文原文

>> No, I think journalists hallucinate.

[笑声]

英文原文

>> [laughter]

但不，它没有任何含义。所以他写了申请。

英文原文

>> But no, it doesn't mean anything. So he wrote the application.

他写了，而我给他的反馈是，嗯，你知道，人们已经尝试过无数次了。法律科技从来没有成过气候，因为律师没有动力去——他们为什么要提高效率呢？他们赚的钱会更少。这根本说不通。但我认为律师确实想要采用它的原因是，他们理解形势的严峻。他们知道商业模式和一切都在面临压力，他们需要做出改变，因为竞争压力太大了。计费模式本身我认为也将会改变。我们已经看到了这一点，特别是在美国，很多工作实际上是按项目收费，而不是按小时收费。嗯，特别是这个光谱上低端的事情，比如不管是什么小型的审查之类的。有些说实话甚至在免费做，有些则只是按固定费率定价。

英文原文

>> He he wrote and my feedback to him was um you know, people have tried this a million times. Legal tech has never been a thing cuz like there's no incentive for lawyers to to why would they be more efficient? They earn less money. Like it doesn't make any sense. But I think the reason that lawyers do want to adopt this is that they understand the gravity of the situation. They know that like the business model and everything is under pressure and it's like they need to change cuz the competitive pressure is so high. The billing model itself I think is also going to change. We've we're already seeing that especially in the US that a bunch of stuff is actually being priced on like a project fee and not on an hours fee. Um particularly this like sort of lower end stuff on the spectrum like whatever it is like small reviews and whatever. Some of it honestly is being done for free and some of it is just being priced on a flat rate.

那钱是从哪里赚的呢？是因为——

英文原文

>> So where is the money being made then? Is it because

在战略性工作中。

英文原文

>> In the strategic work.

对，我感觉你们谈论的很多都是并购（M&A）。比如钱全是在那里赚的。

英文原文

>> Yeah, I feel like a lot of what you guys talk about is M&A specifically. Like that's where all the money is made.

非常现成的例子。对，对，对。诉讼也是，大块工作依然存在，而且非常有价值。所以他们会——律师会把更多时间花在那些地方。客户关系也非常重要。

英文原文

>> very easy sample. Yeah, yeah, yeah. Litigation is also the big work still is there and it's very valuable. And so they'll they'll spend lawyers will spend more of their time there. Client relationships also super important.

就像客户关系管理（CRM）？

英文原文

>> Like a CRM?

不，只是花更多时间陪伴客户。比如很多律所会——或者历史上基本上会为了维持与客户的关系而免费做事。

英文原文

>> No, just like spending more time with the client. Like a lot of law firms will or historically have basically done stuff for free just to have the relationship with the client.

是的，是的。

英文原文

>> Yeah, yeah.

呃，所以现在他们可以那样做，而不必因此损失很多钱。

英文原文

>> Uh and so now they can do that but they don't have to lose a lot of money on doing it.

我还以为——我还以为我必须付钱给律师才能让他们当我的朋友呢。难道不是——

英文原文

>> I thought I thought I just I have to pay the lawyer money just to be my friend. Like isn't

那……那不就是常年法律顾问费那套吗？

英文原文

that

isn't

that

the

retainer

thing?

对，对。对，他们做过那个。那就是……那就是……

英文原文

>> Yeah,

yeah.

Yeah,

they

did

that.

That's

that's

你……你跟我说，你跟我说我可以免费做，比如……

英文原文

>> You

you

tell

me

you

tell

me

I

can

do

it

for

free,

like

对，你只需要找到合适的切入点去聊［笑］。是的。额，好的，挺酷。

英文原文

>> Yeah,

you

just

have

to

find

the

right

way

to

talk

about

[laughter]

it.

Yeah.

Uh

okay,

cool.

或者法律行业必须变得大得多，我不知道。

英文原文

>> Or legal

space

has

to

grow

much

bigger,

I

don't

know.

是的，是的，我们正在努力。我们正在努力。

英文原文

>> Yeah,

yeah,

we're

trying.

We're

trying.

［笑］

英文原文

>> [laughter]

超级 AI 律师助理与法律工程师

额，你知道，你知道，所以好吧，所以所以所以你是说，嗯，你知道，显然战略工作和所有这些高附加值的东西依然存在，但我们的法律费用并没有下降，不过你也说过律师助理没有受到影响，或者说他们只是在转变。这也是我们在 Cognition 那边曾经担忧过的事情，比如人们会担心那种“我们将取代你的初级工程师”的措辞，因为没人喜欢那种说法，但实际上，是的，其中某部分确实发生了。其他初级工程师则接纳了它，并且变得更有成效，正如你刚才所说的那样。所以，我想知道，对于这个角色或这份工作，是否有某种新称谓正在你们的客户群中受到欢迎？

英文原文

>> Uh

you

know,

you

know,

so

okay,

so

so

so

you're

saying

um

you

know,

obviously

the

the

strategic

work

and

all

these

high-value

add

is

still

there,

but

our

legal

fees

are

not

going

down,

but

you

also

said

paralegals

are

not

affected

or

like

they're

they're

just

shifting.

And

and

this

is

something

that

uh

we

also

did

with

uh

worry

about

with

Cognition

like

uh

people

worry

about

the

phrasing

of

like

we

will

replace

your

junior

engineer

cuz

nobody

likes

that

kind

of

statement,

but

in

actual

fact,

yes,

like

some

some

part

of

it

uh

does

happen.

Other

junior

engineers

embrace

it

and

they

become

more

productive

exactly

like

you

just

said.

So,

I

guess

like

it

is

is

there

a

new

term

for

this

role

or

this

this

job

that

is

taking

that

is

having

traction

with

your

customers?

你指的是超级 AI 律师助理？我还没发现有什么现成的称呼。也许我们应该创造一个。

英文原文

>> You

mean

the

the

super

AI

paralegal

Nothing

I

found

yet.

Maybe

we

should

coin

one.

不就是法律……

英文原文

>> Isn't

it

legal

法律工程师？那是我们内部叫法，而且现在也成了一个行业层面的概念。

英文原文

>> Legal

engineer?

That's

our

and

and

also

now

an

industry

thing.

对。

英文原文

>> Yeah.

CSJD（计算机与法律双背景人才）。

英文原文

>> CSJDs

没错。对，对，对。对。

英文原文

>> Exactly.

Yeah,

yeah,

yeah.

Yeah.

也许他们……那些人是最适合做这类事情的。

英文原文

>> And

maybe

that's

the

there

those

other

people

that

are

best

positioned

to

do

this

kind

of

stuff.

我觉得“肉身代理”是个贬义词，意思就是你只是坐在机器前，输入别人让你输入的指令，然后重复机器让你说的话。

英文原文

>> I

think

that

the

the

derogatory

term

of

meat

proxy

where

you

just

like

sit

in

front

of

the

machine

and

type

in

something

that

that

someone

else

tells

you

to

type

and

then

you

repeat

what

the

machine

tells

you

to

say.

是的。

英文原文

>> Yeah.

但我觉得一个非常优秀的“肉身代理”，就像是你非常了解智能体，你非常了解 Leya，并且你知道如何像……这就是我的工具包，你知道，每个行业都有工具包，而我能用我的工具去完成你需要我做的事情。

英文原文

>> But

like

I

do

think

a

very

good

meat

proxy

is

just

like

well

as

you

know

agents

very

well

and

you

you

know

Legora

very

well

and

you

know

how

to

like

like

this

is

my

toolkit

and

you

know,

every

industry

has

a

toolkit

and

I

can

use

my

tools

to

do

the

thing

that

you

need

me

to

do.

但我想这样一来你其实就不算什么代理了，对吧？你实际上只是在做有意义的工作。

英文原文

>> But

I

I

guess

then

you're

not

really

a

proxy,

right?

You're

actually

just

doing

meaningful

work.

是的，这就好比我不再需要……你知道，我以表格审查为例，因为那是非常常见的事情。我有成百、上千、上万、成千上万个……

英文原文

>> Yeah,

it's

it's

just

like

I

no

longer

like

you

know,

well

I'll

talk

about

tabular

review

cuz

that

that's

a

that's

a

thing

that

happens

a

lot.

I

have

hundreds,

thousands,

millions,

thousands.

我想我们最多可以处理多达一百万个单元格。

英文原文

>> We

do

up

to

a

million

cells,

I

think.

在单次、单次……那确实令人印象深刻。额，我们应该聊聊表格和窗口化处理。对，你知道，我手头有一个大案子，我有一百万份文件，我输入进去。通常我会派出一组律师助理，把他们关在房间里……

英文原文

>> In

a

in

a

single

single

That's

impressive.

Uh

we

should

talk

about

the

the

the

the

table

and

the

windowing.

Yeah,

you

know,

I'm

like

I'm

like

I'm

working

on

a

big

case.

I

have

a

million

documents.

I

type

it

in.

Usually

I'll

throw

a

pair

throw

like

a

team

of

paralegals

lock

them

in

他们会坐在那里一直干到凌晨 4:00 左右。

英文原文

>> They'll

sit

there

till

like

4:00

a.m.

像这种事就没必要发生了，对吧？因为现在他们可以直接提取出来，你还可以去对所有内容重新提示。

英文原文

>> Like

that

that

doesn't

need

to

happen,

right?

Cuz

now

they

can

extract

it

and

you

can

go

and

re-prompt

everything.

嗯哼。

英文原文

>> Mhm.

所以是的，我觉得现在我不再需要为此雇用律师助理，但一个优秀的、超级 AI 赋能的律师助理会去干别的事情，你懂吧？

英文原文

>> So

like

yeah,

and

I

think

now

I

don't

need

a

paralegal

for

that,

but

like

a

good

like

super

AI

pulled

paralegal

would

do

something

else,

you

know?

是的，对，没错。不管那是……

英文原文

>> Yes,

yeah,

exactly.

Whatever

that

他们可以做更多的事情，或者把更多时间花在那些他们以前没有时间做的事情上。

英文原文

>> They

can

do

more

stuff

or

like

spend

more

time

on

the

things

that

they,

you

know,

didn't

have

time

for

before.

或者真的能去睡觉，谁知道呢？不。

英文原文

>> Or

actually

sleep,

who

knows?

No.

产品自用与并购实践

我有一个问题。你们收购了一家叫 Kira 的公司。

英文原文

>> I

have

a

question.

So

you

guys

bought

a

company

Kira.

对。

英文原文

>> Yes.

是这么读的吗？而且你们在做那笔并购时使用了自己的产品。

英文原文

>> Is

that

how

you

say

it?

And

you

you

guys

used

your

own

product

to

do

that

M&A.

对，我们一直都这么做。

英文原文

>> Yes,

we

do

that

all

the

time.

那管用吗？能稍微聊聊……

英文原文

>> did

that

work?

Like

can

you

talk

a

little

bit

about

噢，“吃自己的狗粮”。

英文原文

>> Ooh,

dog

fooding.

是的，完全管用，效果很好。我们经常这么做。我们的法务团队……我是说，案件受理（intake）就是和我们的法务团队一起搭建的。因为之前我们有一个 Slack 工作流，大家都把材料往里面塞，他们觉得“这太烦人了”。

英文原文

>> Yeah,

no

that

works

great.

We're

doing

that

a

lot.

Our

legal

team

is

is

I

mean

intake

has

been

built

with

our

legal

team.

Cuz

right

now

we

have

a

Slack

workflow

where

everyone

flows

their

stuff

in

and

they're

like

this

is

annoying.

对，对。

英文原文

>> Yeah,

yeah.

对吧？而且效率低下。所以大家就觉得太好了，我们应该在这件事上使用法律工作流。

英文原文

>> Right?

And

inefficient.

So

it's

like

great,

we

should

use

legal

workflow

for

that.

对，对。能稍微讲讲那个吗？比如那个过程是怎样的？

英文原文

>> Yeah,

yeah.

Can

you

talk

a

little

bit

about

that?

Like,

you

know,

what

that

process

looked

like?

当然可以。其实就是我们自己做了大量的尽职调查，因为我们可以用自己的 Kira View 和其他工具来完成。我们确实仍然会请外部法律顾问介入，来评估某些风险。或者比如涉及薪酬事宜，Kira 还算简单的，因为只涉及瑞典，但你知道，我们在加拿大做过一起，在澳大利亚也做过一起。所以比如澳大利亚的薪酬体系，好吧，我们是否希望有人能从更偏战略的角度来看待这件事？因此在战略层面，我们依然会聘请律师，让他们真正扮演陪练与探讨伙伴的角色，并在如何构思妥善架构方面给予我们很大帮助。但很多比较琐碎的工作，我们实际上完全可以自己搞定。所以早期发现的很多情况……是的，我们很多早期发现的事项，而且我们确实遇到过发现某种隐患警报后叫停踩刹车的情况。

英文原文

>> Yeah,

absolutely.

So

it's

it's

we

do

a

lot

of

the

due

diligence

ourselves

because

we

can

do

it

with

our

Kira

View

and

other

tools.

We

do

still

have

counsel

coming

in

to

like

assess

certain

risks.

Or

like,

you

know,

there's

payroll

and

so

so

Kira

was

simple

cuz

that

was

only

Sweden,

but

you

know,

we've

done

one

in

Canada

and

one

in

Australia.

So

there's

like

payroll

in

Australia.

Okay,

do

we

want

someone

to

look

at

this

like

a

bit

more

strategically?

So

there

we

will

still

use

lawyers

for

the

strategic

part

where

they

can

actually

be

sparring

partners

and

help

us

a

lot

on

like

how

to

figure

out

how

to

structure

this

the

right

way.

But

a

lot

of

the

more

menial

stuff

we

could

actually

just

do

ourselves.

So

like

a

lot

of

the

yeah,

a

lot

of

the

early

things

we

would

find

and

we

have

had

cases

where

we

found

something

that's

been

like

a

red

flag

and

then

we've

we've

pulled

the

brakes.

能稍微聊聊其中的具体细节吗？

英文原文

>> Can

you

talk

a

little

bit

about

any

of

that?

不行，我想不太行。

英文原文

>> No,

I

don't

think

so.

［笑］

英文原文

>> [laughter]

好的。

英文原文

>> Yeah.

美欧律所的采纳与定制差异

好吧。在更早的时候你说过，你的早期客户是瑞典的。

英文原文

>> All

right.

Early

on

you

said

you

your

early

customers

were

Swedish.

嗯哼。比如瑞典的律师事务所。

英文原文

>> Mhm.

Like

Swedish

law

firms.

你是否发现与瑞典律所合作跟与美国律所合作之间存在差异？

英文原文

>> Have

you

seen

a

difference

between

working

with

like

Swedish

law

versus

like

American

law

firms?

嗯哼。

英文原文

>> Mhm.

类似这方面的情况？而且你们大部分是跟非常非常大的大型律所合作，对吧？

英文原文

>> That

sort

of

thing?

And

you

guys

are

mostly

working

with

like

big

big

law

firms,

right?

嗯，我们也有规模较小的律所。我们毫无疑问是从规模非常非常大的大律所起步的，但 50 名律师规模的机构也在使用我们，并且获得了很大价值。

英文原文

>> Well,

we're

also

smaller

law

firms

also.

We

still

we

definitely

started

really

really

big

law

firms,

but

50

lawyers

also

use

us

and

and

get

a

lot

of

value.

是的，那么你有看到与瑞典律所合作和与美国律所合作之间的差异吗？甚至比如我不知道，亚洲的律所。

英文原文

>> Yeah,

so

have

you

seen

differences

between

working

with

like

the

Swedish

ones

versus

American

ones?

Maybe

like

even

I

don't

know,

Asian

ones.

有的，有的，有的。我觉得欧洲的律所……嗯……行动相对要慢一些。美国的律所非常前沿，行动非常迅速，我认为是这样。额，而且这在美国公司的议程上排得非常靠前，我认为比欧洲公司更受重视。这就像是一件具有战略意义、极其重要的事情，我们必须在这方面大干一场。美国公司普遍来说——无论是企业法务部门还是律师事务所——对于定制构建功能也有强得多的意愿。这是他们全都想做的一件事。比如“嘿，我们想跟你们一起打造这个庞大的东西，不管以何种方式去构建”。而欧洲人则更倾向于“我们就用现成的开箱即用方案，我们只要……”。

英文原文

>> Yeah,

yeah,

yeah.

I

think

the

European

ones

um

have

been

a

bit

slower

to

move.

The

American

ones

are

very

forward,

um

move

really

quickly,

I

think.

Uh

it's

also

very

high

on

their

agenda

of

the

American

companies,

more

so

than

the

European

ones,

I

think.

It's

like

a

strategic,

super

important

thing

we

need

to

just

like

do

a

lot

here.

The

American

companies

generally,

both

corporates

and

law

firms,

also

have

a

much

higher

appetite

to

like

custom-built

stuff.

That's

like

a

thing

that

they

all

want

to

do.

Like,

"Hey,

we

want

to

build

this

like

huge

thing

with

you

guys

in

like

structured

in

whatever

way."

Whereas

Europeans

are

much

more

like,

"We'll

just

take

the

off-the-shelf

and

we'll

just

他们想要自己构建的东西有什么例子吗？

英文原文

>> What

's

an

example

of

something

they

would

want

to

build

themselves?

集成到他们自己的系统里，比如其中一些客户构建了自己的 CRM，以及他们自己的……

英文原文

>> Integrations

into

their

own

like

some

of

these

customers

built

their

own

CRM

and

their

own

额。

英文原文

>> Uh

文档管理系统，以及各种他们自己的系统。他们想要构建那个，然后他们想要构建触发机制，比如在 Salesforce 中进行一项审查，出于某种原因流转到某一个人那里，他们做点操作，然后进入核心系统，然后再流转回去。还有……那是比较简单的一端；然后还有一些极其雄心勃勃的，相当于他们想和我们一起构建一个全新的产品，而你知道，我们其实没办法做那个。

英文原文

>> document

management

system

and

their

own

all

sorts

of

stuff.

They

want

to

build

that

and

then

they

want

to

build

triggers

that

like

you

do

a

review

in

in

in

Salesforce

and

it

goes

to

this

one

person

for

whatever

reason

and

they

do

something

and

then

it

goes

into

the

core

and

then

it

goes

back

in.

And

there

are

That's

like

the

simple

end

and

then

there

are

some

that

are

extremely

ambitious,

where

it's

like

a

whole

new

product

that

they

want

to

build

with

us

and

I

you know,

we

can't

really

do

that.

对，对。［笑］

英文原文

>> Yeah,

yeah.

[laughter]

嗯，但那绝对是一个现象，就是美国公司比欧洲公司更想自己去构建。

英文原文

>> Um

but

but

that's

definitely

a

thing

where

like

American

companies

really

want

um

to

build

their

own

more

than

European

ones

do.

那欧洲的客户是那种，比如“嘿，你知道，他们会给你们提产品建议吗”？还是……

英文原文

>> And

the

European

ones

are

like,

"Hey,

like,

you

know,

they

give

you

product

suggestions?"

Or

对，当然会。我是说，我们和他们所有人都有极好的合作。所以他们会说：“这个太难用了。这个太棒了。我们希望更多这种功能。”然后我们就会打造出具备该功能的产品。

英文原文

>> Yeah,

for

sure.

I

mean,

we

have

great

collaboration

with

all

of

them.

So,

they'll

they'll

say,

"This

sucks.

This

is

great.

We

want

more

of

this."

And

we'll

just

build

the

a

product

that

does

that.

好吧。所以他们并没有在自己的业务中看到某种生存威胁。

英文原文

>> Okay.

So,

they

don't

see

like

an

existential

threat

here

in

their

business.

我是说，他们……

英文原文

>> I

mean,

they

反之美国那些律所看到了？还是说……

英文原文

>> Whereas

the

American

ones

do?

Or

我不知道这是否是他们想自研定制而非不自研的原因。我认为双方都知道这极其重要，是他们律所和公司成功的关键。所以两者都认为这极其重要。我认为这只是一种达成目标的不同方式。

英文原文

>> I

don't

know

if

that's

the

reason

that

they

want

to

custom-build

versus

not

custom-build.

I

think

they

both

know

that

this

is

like

uh

extremely

important

and

key

to

their

firm's

success

and

their

corporate

success.

So,

they

it's

both

extremely

important.

I

think

it's

just

a

different

way

to

achieve

it.

好的。另一个问题我想也许你想提出来，那就是产品指标。比如，你怎么知道我们在这方面实际上做得不错？

英文原文

>> Okay.

Another

question

I

think

maybe

you

want

to

bring

this

up

is

like

product

metrics.

Like,

how

do

you

know

that

like,

you

know,

we're

actually

doing

something

good

here?

是的，你对此怎么看？我猜想这也和评测（evals）有关。

英文原文

>> Yeah,

how

do

you

think

about

that?

I'm

assuming

that's

also

related

to

evals.

我们有非常关注的常规产品指标。大家在我们产品中花费的时间多吗？额，他们看起来从中获得了很大价值吗？他们续约了吗？他们经常回访使用吗？作为一款企业级软件，我们有着非常漂亮的数据表现，比如我们的日活/月活（DAU/MAU）以及我们的……比如你……

英文原文

>> So,

we

have

the

regular

product

metrics

that

we

care

a

lot

about.

Are

people

spending

a

lot

of

time

in

our

product?

Uh

do

they

seem

to

get

a

lot

of

value

out

of

it?

Do

they

renew

it?

Do

they

come

back

a

lot?

We

have

really

good

numbers

for

an

enterprise

software

of

like

our

down

over

mile

and

our

like

you

评估指标与客户数据隐私

你知道，用户每月在产品中花费的时间极高。然后我们也会做某种在线邮件（evals），试图大体弄明白：我们在这里解决问题了吗？就类似于用超级廉价的小消息，试图建立一个在线指标，这样在我们汇总发布新功能时也可以用到它。我们可以直接按照特性标志（feature flag）来进行对比，看看是确实改进了，还是出现了灾难性的下滑？因为实际上有一点可能很有趣，对我们的业务来说也很棘手，那就是我们基本上不能查看任何 trace。我们不能看任何数据，因为它们全都是高度机密的。

英文原文

know, hours spent per month in the product extremely high. And then we do some sort of we do online emails also where it's like try to figure out generally like do we solve the problem here? Just like with the super cheap little messages try to just like have an online metric and then we can use that also when we do roll ups of new things. We can just like compare feature flag wise, okay, does it did improve or like did it catastrophically decline? Because one thing actually that's might be interesting that's difficult for our business is we can't look at any traces basically. We can't look at any data cuz it's all super confidential.

在工作中面对这种情况极其令人沮丧，因为你基本上不知道人们到底在做什么。所以情况就像是，这里有一个 Agent，它可以做这么多不可思议的事情，而我完全不知道人们实际上在拿它做什么。

英文原文

And that's like extremely frustrating to work with cuz you basically don't know what people are doing. So it's like here's an agent and it can do so many amazing things and I have no idea what people are actually doing.

我之前还以为你们跟律所签了某种协议，他们就会说，噢没问题，当然可以。

英文原文

>> have assumed you guys like signed some sort of sort of agreement with the law firm and they're like, oh yeah, sure.

是啊，我也希望如此。不，但问题在于，他们拥有的很多数据甚至都不是他们自己的。

英文原文

>> Yeah, I wish. No, but the thing is like a lot of the data that they have is not even even theirs.

是他们客户的。

英文原文

>> It's their clients.

而且高度机密，所以我们看不了。因此我们基本上有一个 evals 团队，试图去推测并近似每个人在做什么，然后将其构建进 evals 中，所以这纯粹是业务中一个超级恼人的难点。

英文原文

>> And it's super confidential so we can't look at it and so basically we have an emails team that's like trying to approximate what we think everyone is doing and like build that into emails that's so that's just like a super annoying kink of the business.

你们能做转述（paraphrase）吗？比如 ChatGPT 有个类似 slash feedback 的功能，会直接完整给出整个 trajectory。但显然那对你们来说是不可能的。不过你们可以用 LLM 来做转述，比如从这个真实的东西生成一个足够接近的合成内容，把所有名字都改掉，所有细节都改掉，但依然保留……

英文原文

>> Can you do paraphrase? So for example chat GPT has like a slash feedback thing that directly fully says it's full trajectory. But obviously that would that is not possible for you guys. But you can use an LLM to paraphrase like from this real thing make a synthetic thing that is like close enough but change all the names, change all the details but like still yeah.

我们不能真的……我是说，这当中总是有一个权衡范围，对吧？如果你一路转述精简到只剩三个值，然后挑选其中一个，那显然没问题。如果你只是把名字去掉，那可能行，也可能不行。所以我们通常会选择最安全的方式。

英文原文

>> We can't really there's I mean there's always a a range right of like if you if you paraphrase all the way down to like three values and you pick one of them then obviously that's fine. If you if you just remove the names that that's maybe maybe not fine. And so like we are um generally pick the the most safe way.

好的。

英文原文

>> Okay.

所以我们尽可能地限制自己，因为你知道……

英文原文

>> So like we we restrict ourselves as much as possible because we you know

绝不能有任何风险。

英文原文

>> there should be no risk at all.

但有些事情是可行的，比如我们知道失败模式大概就是这 10 或 20 种，所以我们可以将它归类到其中一种，这是个很有用的信号。额，我们也可以把它分类到不同业务领域的任务类型中，这样我们就心中有数了。这是我们一直在努力想办法以明智方式去做的事情。

英文原文

>> But there are things where it's like, okay, we know that the failure modes are probably these 10 or 20 so we can classify it into one of those and that's a helpful signal. Uh, we can classify it into the types of tasks in the practice areas also and then like we have an idea. It's something that we're always trying to to figure out how to do in a smart way.

云基础设施与律所的安全接纳

另外在隐私方面，我猜即使在两年前，要让律所接受这些肯定也难如登天。但他们对于把自己的数据发送给 Anthropic 或者 OpenAI 是怎么考虑的？这肯定是一件大事。

英文原文

>> Also like on the privacy front I guess, um, I'm assuming even 2 years ago it must have been like insanely difficult to get the law firms to be okay with this but like how do they think about their data being sent over to Anthropic or you know OpenAI? It must be like a huge thing.

我们的所有基础设施都在 Azure 上。

英文原文

>> All of our infrastructure is on Azure.

是吗。

英文原文

>> Yeah.

额，主要就是出于这个原因，因为当我们刚起步时，律所的态度就像是……

英文原文

>> Um, primarily for this reason cuz like when we started law firms were like

AWS？

英文原文

>> AWS?

不行，绝不可能。[笑声]

英文原文

>> No, no way. [laughter]

律所不信任 AWS？

英文原文

>> Law firms don't trust AWS?

是的，是的。现在情况不一样了。不，当时他们觉得像亚马逊、谷歌……

英文原文

>> Yeah, yeah. It's different now. No, but it they were like Amazon, Google,

[笑声]

英文原文

>> [laughter]

我们不能在任何数据上信任他们。所以我们基本上必须把所有东西都托管在微软上，因为他们已经在使用 365 了。

英文原文

>> we can't trust them with any of our data. So we basically we had to host everything on Microsoft cuz they already have 365.

所以他们接受微软，而现在……

英文原文

>> So they're okay with Microsoft and now

没错，因为他们之前已经为所有的文档、Word 等等做过了安全审查之类的工作，所以在他们看来微软是一家经过批准的供应商。因此我们必须把所有东西都托管在 Azure 上，这在过去几年里发生了变化。现在大家都明白了：好，我们需要上云，我们需要做所有这类事情。额，当时也有更多人在讨论本地部署（on-prem），现在那也完全消失了。大家都明白，你直接在云端做就行，它是安全的，而且可能比你现有的做法更安全。

英文原文

>> Exactly cuz they've already done like whatever security review etc. for all of their documents and all of their word and whatever so like Microsoft is an approved vendor in their mind and so we had to host everything on on Azure and that's changed over the past few years. So like now people understand okay, we need to get on cloud. We need to do all this kind of stuff. Um, there was also a lot more people talking about on prem back then. That's also disappeared completely. People understand that like you should just do it in the cloud and it's and it's safe and it's probably safer than what you're doing.

确实。

英文原文

>> Yeah.

额，所以这改变了很多，但当时确实……

英文原文

>> Um, so that's changed a lot but it was really

顺便说一句，这才过了 3 年，所以实际上就像是……

英文原文

>> And by the way, it's only been 3 years so like it's actually like

这是一次……

英文原文

>> It's a

[笑声]

英文原文

>> [laughter]

是的，是的，这是一个巨大的转变。这太棒了。

英文原文

>> yeah, yeah, it's a huge shift. It's it's it's it's great.

我是说这有利于你们，不过是啊，我本来也没指望……法律行业在过去 10 年里不知怎么地一直对这方面视而不见。

英文原文

>> I mean good for you guys but yeah, I mean I don't expect the like you know the legal firm legal industry has somehow ignored this for the last like 10 years.

额，是的。

英文原文

>> Uh, yeah.

是啊。

英文原文

>> Yeah.

我觉得，我觉得我们需要 AI……

英文原文

>> I think I think we needed that AI

[笑声]

英文原文

>> [laughter]

踢一脚屁股，让每个人都明白我们需要稍微改变一下做事方式了。好的，好的。

英文原文

>> kick in the butt for everyone to understand that we need to change our ways a little bit. Okay, okay.

模型套利与语音转录

那么关于开源模型，现在也变得非常出色了。所以对于这一点，结合……你是怎么考虑的？

英文原文

>> So also then on like open source models now are getting really good. So yeah, how are you thinking about that with the

是的，正如你所指出的，我们只是在考虑带领我们的客户去跟上帕累托前沿（Pareto frontier），把握所有模型的全貌：它们擅长什么、它们的延迟、成本、额、合规治理以及区域可用性。我们如何在这所有维度上进行全面优化？因此，某些开源权重模型在特定任务上表现极佳，那我们就会在那些任务上使用它们。额，有些我们会做后训练，而有些我们就直接使用前沿模型。

英文原文

>> Yeah, we're just thinking about to to your point, um, just bring our clients along on like the Pareto frontier and like the of the the map of all the models and what they're good at and their latency and their cost and their uh governance and and and their regional availability. How do we just optimize across all of that? So some open weights models are great for certain tasks and then we'll use them there. Um, some will post train and and some will just use the frontier models.

是的，你称它为你称它为模型套利。是的，你有不同的风味（flavors）。你在今年年初曾对此做过一些研究，当时你坚定地站在 Anthropic 这边。Grok 也在大幅提升。额，是啊，你想举个综合排名不高、但你想夸一下它表现出奇优秀的模型例子吗？

英文原文

>> Yeah, you called you called it model arbitrage. Yeah, you have different flavors. You you did some research on this where you were firmly entropic at the start of the year. Grok is improving a lot. Um Yeah, you want to give an example of a model that wasn't ranked highly overall, but like you want to shout it out as surprisingly good?

好的，我觉得综合来看不算优秀，但 Gemini——早期的某版 Gemini，是的，在特定类型的转录上极其出色。是的，还有说话人区分（speaker attribution）。所以我们实际上把它用于转录。

英文原文

>> Yeah, I think not overall good, Gemini, one of the early Geminis is yeah, extremely good for certain types of transcription. Yeah, and speaker attribution. So, we use that for transcription actually.

所以你们有……是啊，你们做转录是像 Granola 那样吗？

英文原文

>> So, you have a Yeah, are you guys transcribing like a granola?

我们没有——我们其实大概会做一个类似于会议的产品，但目前这只是语音转文本、文本转语音，就像任何其他具有该功能的应用一样。而且我们实际上也会转录文件。如果你上传庭审录音什么的，我们就会把它们转录成文本，这样我们就可以对其进行所有的分析工作。

英文原文

>> We don't have we actually probably will build like a meeting thing, but this is just like talk to speech, speech to text, text to speech, just like an an um any any other app that has this. And we do actually also transcribe files. If you upload hearings or whatever, we'll we'll then like transcribe them for the men to text and so we can do all of our analysis on it.

没用 ElevenLabs？

英文原文

>> No eleven labs?

出于 DPA 的原因没用 ElevenLabs。同样是因为数据处理协议（data processing agreement）。

英文原文

>> No eleven labs for DPA reasons. Same thing is like a data processing agreement.

我早就该想到，所以……

英文原文

>> I should say so like

我们的数据处理协议中包含的供应商非常少。

英文原文

>> our data processing agreement has very few vendors on it.

是的。

英文原文

>> Yeah.

因为确实得放在第一位严加控制。

英文原文

>> Cuz it's really keep it first.

没错。我们需要对任何数据的去向保持极度严格。所以，额……

英文原文

>> Exactly. We need to keep it super tight on where any data goes. And so um

我觉得不，这也正是像 Gemini 和 OpenAI 这类平台的优势之一：你来到我们这里，你所需的大部分东西我们都有。这实际上非常有价值。即使它不一定是最顶尖的技术，老实说它在大多数任务上通常也已经足够好了。

英文原文

>> I think no, that that is one of the benefits of like a Gemini and Open AI where they're like you you come to us, we have most of the things that you need. And that is actually very valuable. Even if it's like it doesn't have to be like state of the art. It's honestly like mostly good enough for all tasks.

对啊，那谁还在乎呢？

英文原文

>> Yeah, then who cares?

是啊。

英文原文

>> Yeah.

就像……

英文原文

>> Like

是的，[笑声] 不，完全是这样。我其实还应该夸一下微软的转录模型。

英文原文

>> Yeah, [laughter] no, absolutely. I should actually shout out also Microsoft's transcription model.

他们做了一个？My Transcribe？

英文原文

>> They made one. My My transcribe?

是的。

英文原文

>> Yeah.

好吧。

英文原文

>> Okay.

它非常好用。

英文原文

>> It's great.

好的。

英文原文

>> Okay.

它比旧版的 Gemini 更好。

英文原文

>> It's better than the old Gemini.

对，对，对。他们确实取得了一些不错的结果。我都不知道——我觉得通常它们在竞技场（arena）得分上都排在前二、前三、前四，这就足够好了。

英文原文

>> Yeah, yeah, yeah. They they they did some good results. I don't know don't even know like I think like usually all of them are like top two, top three, top four in terms of like the the arena scores and that's good enough.

是的，确实如此。就像是不仅具备所有的治理合规性，扩展性又极高，那这样可能就足够好了。

英文原文

>> Yeah, exactly. And it's like you have all the governance and it's super scalable and it's like then that's probably good enough.

是的，我本来想说，现在语音转文本模型是不是基本上已经足够好了，达到差这么一两个词也无所谓的程度了？

英文原文

>> Yeah, I was going to say like at this point are pretty much the speech to text models like good enough to where you're like one word here or there if it's off, that's fine.

不，不，不。

英文原文

>> No, no, no.

这种事我做过很多次。尤其在法律领域，曾经发生过这种情况，我想是在——我忘了是不是在澳大利亚，法庭上对某个人有一种特定的称呼，结果模型每次都会把它转录成 condom。

英文原文

>> Like I've done this a lot. And particularly in legal, there's like some we we had this thing where I think it's I forget if it's like in Australia where there's like a certain name for like a person in court and it would consistently transcribe it as condom.

[笑声]

英文原文

>> [laughter]

所以对，总会有某些你绝对不希望出现在那里的词。

英文原文

>> So there yeah, there's there's certain things that you don't want there.

是的，但通常你会在其之上运行一层 LLM，然后提取出一些关键词，接着带着这些关键词重新运行转录模型。

英文原文

>> Yeah, but so usually you run an LM layer on top of it and then you you extract some keywords and then you run the transcription model again with the keywords.

这差不多是我目前的处理流程。不过没错，如果你在任何大规模场景下做过这事，只要涉及的领域不是那种……

英文原文

>> Like this is kind of my my workflow so far. But yeah, if you if you've just done this on any large scale, you will pick up tons and tons of transcription errors on any domain that is not like

[嗤笑]

英文原文

>> [snorts]

普通日常对话，就会抓出一堆又一堆的转录错误。

英文原文

>> average day-to-day conversation.

是的，

英文原文

>> Yeah,

[笑声]

当然。如果里面有任何行话——

英文原文

[laughter]

absolutely.

If

there's

any

sort

of

jargon

in

there

对。

英文原文

>> Yeah.

那就会很难办。

英文原文

>> it's

going

to

be

tough.

对，比如我觉得 Whisper 和 11 Labs 仍然会把 Claude Code 转录成 cloud code。

英文原文

>> Yeah,

like

I

think

Whisper

and

11

Labs

still

transcribe

cloud

code

as

cloud

code.

是的。

英文原文

>> Yeah.

而且那种情况——

英文原文

>> And

like

that's

对，我确实，我确实遇到过那种情况。

英文原文

>> Yeah,

I've

I've

actually

had

that

happen.

非常非常普遍。

英文原文

>> very

very

common.

是的。

英文原文

>> Yeah.

所以，所以你就能据此看出是不是还有很多提升空间。

英文原文

>> So

so

like

that's

that's

how

you

know

if

there's

like

a

lot

more

to

improve.

核心产品：表格审查与 Agent

对，确实。我确实想聊一下 Agent 产品这边。我们刚才已经聊了几个高层面的内容，对吧？我们提到了表格审查（tabular review），提到了工单接入（intake），你手头还有一些其他东西我们甚至都还没涉及。但我只想聊聊那两个大件，姑且这么叫吧。说它们是两个大件，这样讲妥当吗？

英文原文

>> Yeah,

absolutely.

I

do

want

to

cover

a

little

bit

on

the

the

agent

product

side.

So

we've

covered

a

few

like

the

of

the

top

level

stuff,

right?

We've

already

named

drop

tabular

review,

we

named

drop

intake

and

you've

you've

got

some

other

stuff

that

we

haven't

even

touched

on

yet.

But

I

just

want

to

cover

the

the

big

two,

let's

call

it

that.

Is

it

is

that

is

that

fair

to

say

they're

the

they're

the

big

two?

Agent 和表格审查。可能还有，是的，还有我们的 Word 插件也是个重点，因为很多人都在 Word 里面工作，所以 Word 插件非常重要，Outlook 插件也是。

英文原文

>> Agent

and

tabular

review.

And

probably

yeah,

and

our

word

add-in

is

also

something

cuz

that

a

lot

of

people

work

work

inside

of

word

and

so

word

add-in

is

is

a

huge

and

an

outlook

add-in

also.

对，所以我就是想聊聊大家都必须要了解的那几个大件。

英文原文

>> Yeah,

so

I

just

want

to

cover

like

the

the

big

ones

that

everyone

must

know.

好。

英文原文

>> Yeah.

所以表格审查基本上就是针对每份文档做大量的扇出（fan out）。你可以跨所有文档提同样的问题，收集到一张表格里，然后来回处理它。处理一百万份令人印象深刻。还有什么其他的工程考量或性能方面的问题吗？比如这里面有什么难点？

英文原文

>> So

tabular

review

is

basically

like

fan

out

of

like

a

lot

across

each

document.

You

can

ask

the

same

question

across

all

documents,

collect

it

in

a

table

and

then

work

back

and

forth

with

it.

A

million

is

impressive.

What's

any

other

like

engineering

considerations

or

performance

things?

Like

what

what's

hard

about

this?

在我们聊那个之前，你能不能先梳理一下律师的使用体验是什么样的，然后比如——

英文原文

>> Really

before

we

do

that,

can

you

just

like

walk

through

like

what

is

the

user

experience

for

a

lawyer

and

then

like

好的，所以你是想看看顶层的流程，是吧？

英文原文

>> Okay,

so

you

want

you

want

to

see

the

top

level

flow,

yeah?

对，对，比如他们用表格审查具体在做什么，然后我们再深入探讨。

英文原文

>> Yeah,

yeah,

like

so

like

what

are

they

doing

like

with

tabular

review

and

then

we

can

go

into

that.

好的。

英文原文

>> Yeah.

所以他们，通常我是说一个典型的例子就是尽职调查，对吧？就是我们需要审查所有的雇佣合同、所有的供应合同、所有的不管什么合同。对于雇佣合同，你把 10,000 份雇佣合同放进去。然后你，呃——

英文原文

>> So

they

it's

typically

I

mean

a

good

example

is

It's

a

due

diligence,

right?

Where

it's

like

we

need

to

review

all

employment

contracts,

all

supply

contracts,

all

of

whatever.

And

so

employment

contracts,

you

put

10,000

employment

contracts

in

there.

And

then

you

um

比如每个单元格是一份雇佣合同吗？

英文原文

>> Like

each

cell

is

one

employment

contract?

每一行是一份雇佣合同。然后对于单元格，你基本上就是提问题。比如第一个问题，第一个，呃不好意思，是按列来。所以第一列可能是：解雇通知期限是多久？然后它就会给所有人填出来。不管是 30 天还是多少。下一个问题比如：是否有知识产权转让条款，诸如此类？再下一个比如：如果解雇他们，你知道，我们得赔付多少钱？或者不管是什么。比如他们的薪资是多少？然后它就会把整张表填完。

你也可以直接让 Agent 来做这件事。比如你只要把 10,000 个文件丢给 Agent，然后说：对此做一次尽职调查，它就会启动一个表格审查并把全部文件跑一遍，等待它完成，然后分析结果，接着它就能高亮显示预警信号（red flags）。好，这位 CEO 拿的薪水高得离谱，或者这里有某种奇怪的奖金结构。好，那显然我们需要看一下。然后 Agent 可以据此起草一份备忘录发送给客户，或者也可以只做审查，然后让你所聘用的律师直接说：好的，我们来看看这个，这些是预警信号。它也是支持协作的，所以你可以在上面发表评论，呃供其他人查看等等。

英文原文

>> Each

row

is

one

employment

contract.

And

then

for

cells,

you

ask

questions

basically.

So

like

the

first

question,

the

first

uh

sorry,

per

column.

So

the

first

column

might

be,

what's

the

termination

uh

period?

And

then

it's

going

to

like

fill

out

for

everyone.

It's

30

days

or

whatever

it

is.

And

the

next

question

is

like,

is

there

an

IP

assignment

clause

or

whatever?

Next

one

is

like,

how

much

you

know,

how

much

do

we

have

to

pay

them

if

we

fire

them?

Or

whatever

it

is.

Like

how

much

are

they

getting

paid?

And

then

it's

going

to

fill

out

the

entire

thing.

You

can

also

just

ask

the

agent

to

do

this.

Like

if

you

just

drop

10,000

files

into

the

agent

and

say

like,

do

a

due

diligence

on

this,

it

will

spin

up

a

tabular

review

and

run

all

of

them

and

wait

for

it

to

be

done

and

then

analyze

the

results

and

it

can

then

highlight

like

red

flags.

Okay,

this

the

CEO

is

getting

paid

an

outrageous

amount

or

there's

like

this

weird

bonus

structure.

Okay,

then

like

obviously

we

need

to

look

at

that.

And

then

the

agent

from

that

can

draft

a

memo

to

send

to

the

client

or

uh

can

just

do

the

review

and

then

have,

you

know,

a

lawyer

that

you

use

to

just

be

like,

okay,

let's

let's

look

at

this.

Here

are

the

red

flags.

It's

also

collaborative

so

that

you

can

comment

on

it

um

for

other

people

to

look

at,

etc.

那你们为什么选择电子表格这种形态呢？

英文原文

>> And

why

did

you

choose

that

like

uh

spreadsheet

format?

是的。就是用起来、在里面工作要好得多，超级容易上手。呃非常容易超快速地对当前情况有一个全局概览，并且进行审查。而且我们生成的所有内容都带有追溯到源文档的引用。所以你基本上审查起来超级容易。如果你想确保万无一失，你还可以给单元格做高亮标记。这些是已审查过的，或者不管怎样。这些是被标记出来的，你可以发表评论和进行协作。这比我们尝试过的任何其他产品形态都要容易得多。

英文原文

>> Yeah.

It's

just

a

lot

better

to

it's

it's

super

easy

to

work

with

work

in.

Uh

it's

really

easy

to

get

an

overview

super

quickly

of

what's

going

on

and

to

review

it.

So

all

of

the

stuff

that

we

produce

also

has

citations

back

to

source

documents.

And

so

you

can

basically

review

it

super

easily.

If

you

want

to

just

make

sure

that

everything's

right,

you

can

also

like

highlight

cells.

These

are

reviewed

or

whatever.

These

are

the

ones

that

are

flagged

and

you

can

comment

and

collaborate.

That's

much

easier

than

any

other

form

factor

that

we've

tried.

是的。我想说电子表格这种形态在某些领域吸引力很大，但不知为何并没有真正铺开。我觉得这很奇怪。比如 GTM 领域的 Clay 是另一个你熟知的、以电子表格闻名的产品。你们这家——我真不知道还有谁做大规模的电子表格 AI 产品。

英文原文

>> Yeah.

I

would

say

that

the

spreadsheet

form

factor

has

has

a

lot

of

pull

in

some

areas

but

somehow

hasn't

really

spread.

I

I

I

find

it

very

weird.

Like

Clay

is

another

one

on

the

GTM

side

that

you

know,

very

known

for

the

spreadsheets.

You

guys

I

don't

really

know

if

anyone

else

does

a

large

spreadsheet

AI

thing.

没有。没有。

英文原文

>> No.

No.

之前有比如 V7 试图推出过类似的东西，还有各种表格类产品，甚至在金融领域我都没有什么顶级大牌在做电子表格……

英文原文

>> There

there

was

like

V7

that

like

tried

to

launch

that

like

there's

like

sheets,

there's

even

in

finance

domain

I

don't

really

have

like

top

name

that

like

does

spreadsheet.

做得非常出色的。对。

英文原文

>> really

well.

Yeah.

[笑声]

英文原文

>> [laughter]

我赞同。我认为如果你的场景存在扇出，并且你在做完全相同的事情，你在提取内容，那么对于理解当前状况来说，这真是一个绝佳的产品形态。

英文原文

>> I

I

agree.

I

think

it's

like

a

great

if

you

if

you

have

fan

out

and

you

do

the

exact

same

thing

and

you're

extracting

stuff

it's

it's

such

a

great

form

factor

to

like

understand

what's

going

on.

你觉得用户是非得导出到 Excel，还是说大家很乐意就留在你们的电子表格里？

英文原文

>> Do

you

find

like

you

must

export

to

Excel

or

like

people

are

happy

to

live

in

your

spreadsheet?

大家很乐意就留在那里。

英文原文

>> People are

happy

to

live

there.

呃，你们是用 AG Grid 还是自己写的——

英文原文

>> Uh

do

you

use

AG

Grid

or

your

own

用 AG Grid。

英文原文

>> use

AG

Grid.

AG Grid。那看起来是最领先的一个——

英文原文

>> AG

Grid.

That

seems

to

be

like

the

leading

one

that

我是说他们做这个已经做了大概我不知道 15 年了。

英文原文

>> I

mean

they've

done

this

for

like

I

don't

know

15

years.

是的，是的，他们懂怎么做，对。

英文原文

>> Yeah,

yeah,

they

know

how

to

yeah.

是的，所以，我是说电子表格这个切入点我很满意。我觉得对我而言工程层面唯一另一个有趣的事情就是扇出。

英文原文

>> Yeah,

so

so

yeah,

I

mean

I'm

happy

on

that

that

spreadsheet

angle.

I

mean

I

think

the

the

only

other

interesting

thing

for

me

engineering

wise

is

the

fan

outs.

没错。

英文原文

>> Yep.

工程师也许能够很好地理解权衡取舍，比如好吧，在你把这东西分发到一百万份文档之前，也许先拿两份测一下。

英文原文

>> Engineers

maybe

can

are

well

equipped

to

understand

the

trade

offs

of

like

well,

okay,

before

you

you

send

this

thing

out

on

a

million

documents

maybe

test

it

on

like

two.

嗯。

英文原文

>> Mhm.

呃——

英文原文

>> Uh

[笑声]

英文原文

>> [laughter]

对对，在搞大之前先看看行不行得通？

英文原文

>> Yeah,

yeah,

does

that

work

before

you

go

nuts?

他们会这么做吗？

英文原文

>> Do

they

do

that?

会的，他们会。所以我们的运行类型实际上会先跑前 10 行。

英文原文

>> Yeah,

they

will.

So

our

type

of

the

row

actually

will

run

the

first

10

rows.

对，先确认一下，对吧？

英文原文

>> Yeah,

just

check,

right?

对，看看这说得通吗？

英文原文

>> Yeah,

does

this

make

sense?

看看它显示的结果是不是我们预期的那样？

英文原文

>> So

did

it

did

it

show

us

what

we

thought

it

was

going

to

show

us?

对。

英文原文

>> Yeah.

那挺好的。噢，而且当人们构建这个的时候，让他们在前 10 行上快速迭代也非常合适，直到他们拿到想要的结果，然后他们就可以火力全开。

英文原文

>> That's

a

that's

really

good.

Oh,

but

also

for

people

when

people

built

this

it's

just

really

good

for

them

to

iterate

quickly

on

the

first

10

rows

until

like

they

have

a

they're

getting

the

results

that

they

want

and

then

they

can

go

ham.

工单接入与任务流整合

好的。我很乐意进入下一个话题——工单接入，也就是你们的 Slack 机器人。大家都有 Slack 机器人。你们的有什么不同？

英文原文

>> Yeah.

I

am

happy

to

move

on

to

the

next

thing

intake

which

is

your

Slack

bot.

Everyone

has

a

Slack

bot.

What's

different

about

yours?

我们这个就是法律专用的，它运作的方式其实不是——

它可能也会，我们可能也会做一个纯粹作为 Agent 交互界面的机器人，但它实际上专门用来触发一个 Agent，然后该 Agent 会将一个任务添加到我们的列表（list）中。我们有这个列表功能，它肯定就像是 Linear——

你知道，那是最近似的对标产品。所以它会把它添加到一个特定的接入列表中，然后律师和 Agent 就可以进入这个战局，直接看：好，这些是别人发给我们律所或大企业法务团队的任务。我们可以为它们触发 Agent，可以在上面协作，然后我们直接标记为已完成，接着它就会在 Slack 里回复过去。所以它更像是一个——它不仅仅是触发一个 Agent，它更像是一个，呃——

我不知道，像个门面/前台一样，比如：“嘿，这是工单。我想审查这个东西。”好的，现在完成了，返回去了。谁都不需要改动什么。法务团队只需留在 Legora 里面，在那里管理 Agent 以及所有事务。

英文原文

>> So

ours

is

just

legal

specific

and

what

what

happens

with

ours

is

it's

not

actually

it

might

also

we

will

probably

also

build

one

that's

just

an

interface

to

the

agent,

but

it's

actually

specifically

to

basically

kick

off

an

agent

that

then

adds

a

task

to

our

list.

So

we

have

this

list

thing

which

is

like

linear

for

sure

you

know,

that's

the

closest

equivalent

and

so

it

will

add

it

to

a

specific

intake

list

and

then

lawyers

and

agents

can

go

into

the

war

and

just

be

like

okay,

these

are

the

things

that

you

know,

people

are

sending

to

us

as

our

law

firm

or

as

our

legal

team

in

a

big

corporate.

We

can

kick

off

agents

for

them.

We

can

collaborate

on

them

and

then

we

can

just

like

mark

them

as

done

and

then

it

gets

sent

back

in

Slack.

So

it's

more

of

a

it's

not

just

like

kicking

off

an

agent

it's

more

of

a

like

a

um

I

don't

know

storefront

thing

where

it's

like

hey,

here's

the

ticket.

I

want

this

thing

reviewed.

Okay,

now

it's

now

it's

here

it

is.

It's

back.

No

one

needs

to

change

anything.

The

legal

team

just

sits

in

Legora

and

like

manages

the

agents

and

everything

there.

我觉得这种事情的数据结构非常有趣，因为它基本上就像收件箱，对吧？汇集所有这类的查询，这些查询有可能转化成一件事物，再转化成另一件事物，再转化成别的东西。呃，比如我不知道把所有这些同其他全部功能集成起来很困难吗？还是说你们根本不用担心这个？

英文原文

>> I

think

that

the

data

structure

for

this

kind

of

thing

is

really

interesting

cuz

it's

like

basically

the

inbox,

right?

For

all

these

kinds

of

queries

that

can

potentially

translate

into

a

thing

into

something

else

into

something

else.

Um

and

like

I

I

don't

know

is

it

hard

to

integrate

all

these

with

all

the

other

features

or

do

you

just

not

worry

about

it?

不难。我觉得我们有一些效果非常好的基础原语。比如列表就是一个绝佳的原语，因为本质上它极其灵活。它也像 Notion 数据库。对，Notion 数据库。它就像 Notion 数据库，所以你基本上可以根据任何你想要的来配置它。我们当然投入了大量精力让它具备超高扩展性并且运转得非常好。你可以在上面设置触发器、自动化操作，它就像一个绝佳的粘合剂，把所有东西粘合在一起。因为一旦发生某件事，你就可以基于列表触发一次 Agent 运行，然后它返回并填充某个字段，或者不管怎样添加新文档，或者不管是什么——

英文原文

>> It's

No,

it's

not.

I

think

we

have

some

primitives

that

work

really

well.

So

like

lists

is

a

great

primitive

cuz

it's

essentially

it's

like

extremely

flexible.

It's

also

like

a

notion

database.

Yeah,

a

notion

database.

It's

like

a

notion

database

and

so

you

can

basically

configure

it

for

whatever

you

want

and

we

of

course

invest

a

lot

into

making

this

super

scalable

and

work

really

well

and

you

can

have

triggers

on

them,

automations

on

them

and

it's

like

a

great

glue

to

glue

everything

together

cuz

then

you

can

kick

off

an

agent

run

based

on

a

list

if

something

happens

and

then

that

comes

back

and

populates

a

field

or

whatever

adds

new

documents

or

whatever

it

核心原语与前向部署

本质。就像你可能会用 Linear 或某种任务管理工具来协调一群编程智能体一样，我们也可以用列表来做类似的事。因此智能体可以创建列表，列表可以创建智能体，智能体也可以读取列表，这样智能体集群就鲜活运转起来了。

英文原文

is.

And

so

it's

like

just

like

you

might

coordinate

a

bunch

of

coding

agents

with

linear

or

some

sort

of

task

management

tool.

We

can

do

it

similarly

with

lists

and

so

like

agents

can

create

lists

and

lists

can

create

agents

and

agents

can

read

lists

and

so

it's

like

the

agents

swarm

comes

alive.

[笑声]

英文原文

[laughter]

对，大家对此是感到兴奋还是不知所措呢？你看，我觉得这基本上就相当于终端用户编程了。你可以让用户设置极其复杂的自动化流程。有些人对此超级着迷，甚至成了……就像你懂的，如果能有类似 Legora 的……

英文原文

Yeah,

do

people

get

excited

or

overwhelmed

by

that?

You

know,

like

I

I

feel

like

this

is

basically

end

user

programming.

You're

you

know,

you

can

let

people

set

up

crazy

automations.

Some

people

get

super

obsessed

by

it

and

they

become

like

influent

like

you

know,

it'd

be

cool

to

have

like

Legora

like

Play 平台有网络红人。对，是的。

英文原文

Play

has

influencers.

Yeah,

yeah.

嗯，把事情搞这么复杂值得吗？

英文原文

Um

is

that

worth

it?

Um

to

overcomplicate

it?

即便我们实际上不把这种灵活性开放给终端用户，我们自身也需要它。假设这完全只作为内部使用，那它也是一种能让我们充分借助像法律工程师这样的人群极快构建大量产品的高效途径。例如你可以构建定制化的基金管理仪表盘。你基本上可以在列表的基础上，配合自己的一套生成式 UI 来构建它，然后你基本上就有了一款部分人会非常喜爱的新产品。我们可以打造一个汇聚各种专业化产品的市场空间，并且能以极快的速度扩展规模，因为我们所有的法律工程师都可以随时去构建新的。如果他们在外拜访客户，对方说：“噢，你们能做一个针对我们自己需求的吗？”轻而易举，我们现场就能给你做出来一个。所以这就有点像是在 Legora 内部嵌入了一个 Lovable。

如果我们不想把它对外暴露，我们也是通过这种方式来构建我们自己的定制功能的。我们在英国完成了另一笔收购，那家公司叫 Wexler，他们做的是事实情报分析，主要用于诉讼。比如你针对某起诉讼之类的案件上传了一百万份文件，然后我们从中提取出包含命名实体、日期等信息的事实。当你掌握了所有事实后，你基本上就可以将它们互相核对，看看哪些是得到了佐证的，或者哪些是——你会怎么称呼那种分歧？大家都懂的。这极其强大，因为随后你还可以把法庭听证会的录音转录出来，你就会发现：“噢，他说了这件事，但实际上这边的邮件却显示他当时根本不在场。”如果你现场亲眼目睹这一幕，简直酷毙了。而这一切我们同样可以构建在这个原语之上。因此，基本上我们能依托这一个原语构建出大量非常酷炫的东西。

这就是我们的思考方式，因为随着公司规模不断扩大，我们必须解决太多的问题。所以我会反复思考：我们该如何打造一个核心原语，既能以此扩展出众多不同的终端产品，又能创造出高度可定制的体验。

英文原文

We

need

the

the

flexibility

even

if

we

don't

actually

expose

it

to

the

end

user.

Let's

say

this

was

just

all

internal,

then

it's

a

really

fast

way

for

us

to

utilize

people

like

people

like

our

legal

engineers

to

basically

build

a

lot

of

products

extremely

fast.

So

like

you

can

build

custom

like

fund

management

dashboards.

You

can

basically

build

that

on

list

with

like

your

own

a

generative

UI

and

then

you

basically

have

a

new

product

that

some

people

are

going

to

love

and

we

can

have

a

market

space

of

just

like

a

bunch

of

specialized

products

that

we

can

scale

super

quickly

cuz

all

of

our

legal

engineers

can

just

build

new

ones.

If

they're

out

with

a

client

oh

you

want

your

own?

Easy.

Like

we'll

just

fix

one

for

you.

So

it's

like

a

lovable

inside

Lagora

a

little

bit.

And

if

we

don't

want

to

expose

it,

that's

also

how

we

build

our

own

custom

feature.

So

we

we

made

another

acquisition

in

the

UK

called

Wexler,

and

they're

doing

fact

intelligence,

which

is

for

litigation

mostly,

which

is

like

you

upload

a

million

documents

there

for

a

lawsuit

or

whatever.

And

then

we

extract

facts

from

those

with

like

named

entities

and

dates

and

whatever.

And

then

when

you

have

all

the

facts,

you

can

basically

hold

them

up

against

each

other

and

see

which

ones

are

corroborated

or

which

ones

are

um

what

do

you

call

disagreement?

They

you

know.

And

that's

extremely

powerful

cuz

then

you

can

also

transcribe

a

court

hearing

and

you'll

be

like,

"Oh,

he

said

this

thing,

but

actually

the

email

over

here

says

that

he

was

not

there."

Uh

and

it's

like

if

you

see

it

live,

it's

like

so

cool.

And

that

we

can

also

build

on

this

primitive.

And

so

like

we

can

build

a

lot

of

really

cool

things

on

this

one

primitive,

basically.

That's

how

we

think

about

it

cuz

like

we

have

to

be

as

we

scale

the

company,

we

just

have

to

solve

so

many

things

and

like

so

so

I

think

a

lot

about

how

do

we

build

a

core

primitive

that

we

can

scale

to

to

many

different

end

products

and

create

very

customizable

experiences.

好，那么你们有哪些核心原语呢？我们不妨给听众重新梳理一下。列表算是一个核心原语。

英文原文

Okay,

so

you

have

uh

what

what

core

primitives

do

you

have?

Let's

just

restate

for

for

people.

Lists

is

a

core

one.

是的，我们可以顺着你的说。对，项目——本质上是文件容器、文件系统；嗯，当然还有智能体。你可以启动它，并在各种截然不同的配置下运行它。

英文原文

Yeah,

so

we

can

take

your

one.

Yeah,

projects

basically

containers

of

files,

file

systems,

um

the

agent

of

course.

You

can

kick

that

off

and

run

that

in

very

different

configurations.

顺便问一下，你们给这个智能体起名字了吗？

英文原文

Do

you

have

a

name

for

the

agent,

by

the

way?

没有，我们没起。

英文原文

No,

we

don't.

没起，好的。

英文原文

No,

okay.

没有。

英文原文

No.

好吧。

英文原文

Okay.

对。不，你的意思是我们有没有像叫它 Leia 这样给它取个名字？

英文原文

Yeah.

No,

you

mean

like

we

called

it

like

Leia

or

like

we

gave

it

a

name?

噢，对，Leia。

英文原文

Ooh,

yeah,

Leia.

是啊，叫 Leia 会挺酷的。对。以前大家谈论 Leia 时就像在谈论一个命名实体一样。

英文原文

Yeah,

Leia

would

be

cool.

Yeah.

People

used

to

talk

about

Leia

as

like

an

a

named

entity.

是的。

英文原文

Yeah.

但后来我们就停用了那个叫法。

英文原文

But

yeah,

we

we

stopped

that.

好吧，其实也可以重新拿来用用看。不过可能会引起混淆，我也不确定。

英文原文

Okay,

we

might

as

well

bring

it

back.

It

might

be

confusing.

I

don't

know.

那么，列表、智能体……

英文原文

So

lists,

agent

呃，表格化审查。

英文原文

Uh

tabular

review.

表格化审查。

英文原文

Tabular

review.

我们也在考虑，这到底算不算是……你可以按照自己的意愿尽可能往下细分，对吧？但那些大概就是主要的核心原语了，是的。

英文原文

Which

like

we're

considering

is

that

actually

an

instantiation

of

like

you

can

take

this

as

far

down

as

you

want,

right?

But

those

are

probably

the

big

the

big

ones,

yeah.

好的，那我其实还想聊聊前向部署工程师（FDE），因为听上去有点像那回事。我想 Max 之前也稍微谈过一点。就是说，当你引入一个新的大型客户时，你希望尽快为他们创造价值，这样他们才会继续留在 Legora 平台上。对，这一块是如何演变的呢？你们是什么时候建立起那支前向部署团队的？

英文原文

Yeah,

so

I

actually

wanted

to

talk

a

little

bit

about

like

FTEs

cuz

this

sounds

kind

of

like

that.

And

I

think

Max

has

talked

about

that

a

little

bit.

That

like,

okay,

when

you

onboard

a

new

new

like

big

customer,

you

want

to

get

them

value

like

ASAP

so

that,

you

know,

they

stay

on

stay

on

Legora.

Yeah,

how

has

that

evolved?

Like

when

did

you

guys

start

that

forward

deployed

team?

我们聘请第一名前向部署工程师，那是我们第一次进军纽约的时候，对方也是一位 YC 创始人。大概是 9 个月前？我其实不太确定了。

英文原文

So

we

hired

our

first

forward

deployed

engineer.

That

was

our

first

time

in

New

York,

also

YC

uh

founder.

9

months

ago,

maybe?

I'm

not

actually

sure.

他们算不算是法律工程师呢？

英文原文

Are

they

considered

legal

engineers

or

not?

不算。

英文原文

No.

好的。

英文原文

Okay.

是货真价实的工程师。但我们确实……这也是我一直在思考的事情，那就是：我们究竟该如何规模化地推行这一模式？嗯，而且我认为一个极其核心的要点在于：我们如何才能赋能我们的法律工程师？我们该如何赋能高级用户和我们的法律工程师，让他们能够以强大的方式去定制产品？

当然，我们依然会有前向部署工程师来提供协助，完成所有定制集成、需求范围界定、弄清楚哪些是可行的，并将所有这些反馈回核心产品的开发之中。所以我对此的看法大致是，这就像一道第一道防线和第二道防线：法律工程师应该能够定制大量内容并打造出极佳的体验；而在他们无能为力、或者我们需要对核心产品做出修改、亦或是需要进行定制集成等情况下，我们有前向部署工程师在后方轮转介入——有时他们只负责做产品，有时则会去跟客户打交道。

英文原文

True

engineers.

But

we

do

it

is

something

that

I

think

about

is

like

how

do

we

exactly

enable

that

at

scale?

Um

and

I

think

a

a

really

core

thing

is

like

how

can

we

empower

our

legal

engineers?

Like

how

can

we

empower

our

power

users

and

our

legal

engineers

to

customize

the

product

in

a

powerful

way?

And

then

of

course

we

we'll

still

have

forward

deployed

engineers

that

help

out

you

know,

do

all

the

custom

integrations

and

the

the

scoping

and

figure

out

what's

possible

and

feed

all

of

that

back

into

core

product

development.

So

that's

kind

of

how

I

think

about

it

is

like

almost

a

first

line

second

line

defense

where

the

legal

engineers

should

be

able

to

customize

a

lot

of

stuff

and

build

a

great

experience.

And

then

where

they

can't

or

we

need

to

make

core

product

changes

or

custom

integrations

or

whatever

we

have

forward

deployed

engineers

who

also

then

cycle

in

a

little

bit

behind

like

sometimes

they'll

just

work

on

the

product

and

sometimes

they'll

work

with

the

clients.

你能举个具体的例子吗？比如，你们刚刚签下了一个新客户。第一步是什么？你们会跟对方说“嘿，我们为你们配备了一名前向部署工程师，他们要去为你们打造某种产品”吗？这个过程到底是怎么运作的？

英文原文

Can

you

walk

through

like

an

example?

Like

okay,

so

you

just

got

you

just

got

this

new

client.

Like

what's

step

number

one?

Like

are

you

guys

like

okay,

hey

we

have

a

forward

deployed

engineer

for

you

guys.

They're

going

to

go

build,

you

know,

some

kind

of

product

for

you.

Like

how

does

that

work?

呃，客户与客户之间差别非常大，对吧？但是，如果是一家大客户，他们意愿极高，并且满怀设想、有着各种想做的事，或者已经在做了。通常他们往往已经尝试过自己去构建某些东西，但失败了，他们会说：“这就是我们想解决的问题。”

那么，我们基本上就会视情况为他们配备法律工程师，视具体需求可能还会配备一名前向部署工程师。然后我们就会直接告诉对方：“嘿，这些人会入驻，他们会让你们取得成功。他们会跟你们所有的律师坐在一起，教他们如何使用，摸清你们目前拥有的所有工作流程，并将其构建到 Legora 中。”其中一部分工作仅由法律工程师就能搞定，而另一部分则需要前向部署工程师的参与。

英文原文

Uh

so

it's

very

different

client

to

client,

right?

But

um

if

it's

a

big

client

and

they're

super

engaged

and

they

like

want

they

have

all

these

dreams

of

things

that

they

would

want

to

do

or

they

already

have.

Typically

they

will

have

already

tried

to

build

something

and

it's

failed

and

they're

like

this

is

what

we

this

is

a

problem

that

we'd

like

to

solve.

Um

then

we'll

we'll

basically

staff

them

with

legal

engineers

and

a

forward

deployed

engineer

potentially

depending

on

what

you

know,

their

thing

is

and

then

we'll

just

be

like,

"Hey,

these

guys

will

come.

They'll

make

you

successful.

They

will

sit

with

all

your

lawyers

and

teach

them

how

to

use

it

and

like

figure

out

all

the

workflows

that

you

have

right

now

and

and

build

them

into

Legora."

And

some

of

that

will

just

be

a

legal

engineer

and

some

of

that

will

require

a

forward

deployed

engineer.

我想问一下，这两者之间的界限究竟是什么？能否稍微展开谈谈。

英文原文

What

is

that

separation,

I

guess?

Uh

if

you

could

like

talk

about

that

a

little

bit.

法律工程师是以高级用户的方式来配置产品，他们不写代码。

英文原文

Legal

engineers

configure

the

product

as

a

power

user

would.

They

don't

write

code.

明白了，对。

英文原文

Yeah,

yeah.

而前向部署工程师则基本上是直接向生产环境写代码。我想这就是区分的界限所在。

英文原文

And

a

forward

deployed

engineer

writes

code

into

production,

basically.

I

guess

that's

the

that's

the

the

这些前向部署工程师通常具有什么样的背景？你刚才提到了 YC 创始人。

英文原文

What

kind

of

background

did

the

forward

deployed

engineers

normally

have?

You

said

YC

founder.

对，创始人、正规的工程师。

英文原文

Yeah.

Founders,

regular

engineers.

经验都很丰富。

英文原文

They're

old.

是的。

英文原文

Yeah.

D 就是你们的……

英文原文

D

was

your

没错，D 是我们的第一位。是的。

英文原文

Yeah,

exactly.

D

was

our

first.

Yeah.

做过两家 YC 公司，还在 Meta 做过搜索。

英文原文

Two

YC

companies

and

worked

on

search

at

Meta.

对，她太出色了。是的，没错。

英文原文

Yeah.

She's

awesome.

Yeah.

Yeah.

好的，这门槛可真高。

英文原文

Okay,

that's

a

high

bar.

她非常棒。是的。[笑声] 确实。是的。她是……是的，我们在美国前线的第一批开拓者。

英文原文

She's

great.

Yeah.

[laughter]

Yeah.

Yeah.

Yeah.

She's

Yeah,

first

first

boots

on

the

ground

in

the

US.

是的。

英文原文

Yeah.

那么，这些人通常在帮客户构建什么样的具体工作流程或功能？能否举几个例子？比如你们有没有发现：“噢，这几个用例几乎所有律所都会提到，我们前向部署团队一进驻首先就要去把这些做出来”？

英文原文

So,

what

are

some

examples

of

of

I

don't

know,

workflows

or

things

that

these

people

are

building

for

the

clients?

All

right.

Do

you

guys

have

like

Oh,

these

are

like,

you

know,

there's

a

couple

use

cases

that

like

all

the

law

firms

are

like,

you

know,

this

is

the

first

thing

that

our

FDs

are

going

to

do

when

they

when

they

go

in.

差异很大。嗯，有些核心工作流程大家都有，我们可以直接在 Legora 里构建，那可以是任何事情，比如“我们如何处理这些类型的保密协议（NDA）”之类的。而那只要像刚才说的那样，激活它就行了，对至少法律工程师来说，这其实更像是在构建具体技能。

而对于大多数前向部署工程师而言，更多的是遇到我们目前还做不到的事情。比如我们可能需要在某些原语上获得更多灵活性，或者不管具体是什么，亦或是对方刚好有一个极其复杂的系统需要我们去完成集成对接。于是他们就会去到现场，针对我们的 API 来构建集成方案，并可能在这个过程中扩展我们的 API 等等，以确保方案能够跑通。

英文原文

It's

very

different.

Um

some

core

workflows

everyone

have

and

we

can

just

build

in

Legora

and

that

could

be

anything

from

like

um

this

is

how

we

handle

these

types

of

NDAs

or

whatever.

And

then

it's

just

like

again,

you

activate

it

and

it's

more

like

skill

building

really

for

at

least.

And

most

of

the

FDs,

that's

more

like

there's

something

that

we

can't

do

yet.

Like

we

need

some

more

flexibility

in

some

of

the

primitives

or

whatever

it

might

be

or

we

just

have

this

really

complex

system

that

we

need

to

integrate

into.

And

then

they'll

go

out

there

and

they'll

build

the

integration

against

our

APIs

and

maybe

expand

our

APIs

or

whatever

to

make

that

work.

那么在法律架构层面上又是如何运作的呢？因为你说过你……

英文原文

And

how

does

the

like

legal

structure

work

there?

Because

you

said

you

前线部署工程师与客户端项目

你们自己甚至都不能经手交易。

英文原文

guys yourselves can't even work on deals.

对。

英文原文

>> Yes.

呃，是的。

英文原文

>> Um yeah.

那么，他们是怎么做客户项目的呢？

英文原文

>> So, like how how do they work on client projects?

你是指律所他们怎么做客户项目，还是我们的 FD——

英文原文

>> You mean the law firms, how do they work on client projects or how do our FDs

抱歉。你们的 FD 如果……他们是怎么做客户项目的？

英文原文

>> Sorry. How do your FDs work on client projects if

所以，他们依然不接触数据。他们就只是搭建系统。嗯，但他们有时也会像员工一样办理入职。比如他们可能会领到一台笔记本电脑之类的所有东西。

英文原文

>> So, they don't touch the data still. So, they'll just build the system. Um but they will also be on-boarded like an employee sometimes. So, like they might get a laptop and everything.

比如入职到律所？

英文原文

>> Like to the law firm?

对。对。如果有需要的话。

英文原文

>> Yeah. Yeah. If they need to.

有意思。

英文原文

>> Interesting.

这种情况并不经常发生，因为通常他们只需构建一个集成，大致就能跑通了。不过是的，如果有需要，他们是可以这么做的。

英文原文

>> That doesn't happen very often cuz like usually they just build an integration and like kind of works. But they yeah, they can if they need to.

而且我猜，你们只会在客户达到一定规模时，才会把 FD 入驻进去。

英文原文

>> And you only onboard FDs into a customer, I'm assuming, if they're a certain size.

对。对。

英文原文

>> Yeah. Yeah.

好的。

英文原文

>> Okay.

这得合理才行。因为我们的 FD 也可以做核心产品，对吧？所以这是一个持续权衡优先级的事情。

英文原文

>> It needs to make sense. Cuz our FDs can also work on core product, right? So, it's like it's a constant prioritization thing.

有没有某种规则，比如像大家所知的那样，Palantir 著名的模式是花 4 天时间驻场，1 天时间回来回馈我们的产品。有类似这样的机制吗？

英文原文

>> Is there a rule like I guess does every like, you know, so it's Palantir famously had like, you know, 4 days you spend on site, 1 day you come back and contribute back to our products. Anything like that?

没有。

英文原文

>> No.

我们没有。

英文原文

>> We don't.

坦白跟你们讲，目前在很多方面仍然处于早期阶段。

英文原文

>> I mean, to be fully honest with you guys, still it's still uh you know, it's still early days in a lot of this.

那你觉得未来一年大概会走向何方？

英文原文

>> And where do you see it headed maybe over the next year?

我确实认为它会变得更——也许不完全像 Palantir，但我觉得我们会有更大规模、处于持续轮转中的 FD 团队，也许在某个客户那里驻扎两个月左右来搭建某样东西，然后再去下一个客户，等等等等。是的，但我确实设想过我们带着法律工程师和前线部署工程师进场，去攻坚特定的项目。

英文原文

>> I do think that it's going to be more maybe not exactly like Palantir, but like we I think we will have a larger in constant circulation group of FDs that maybe sit with a client for like two months or whatever to build out this one thing and then they go to the next one etc. etc. Yeah, but I do imagine us rolling in with legal engineers and forward deployed engineers to like work on specific projects.

嗯哼。

英文原文

>> Mhm.

是的。

英文原文

>> Yeah.

API 策略与 Agent 角色

就像你说的，API 业务其实很有意思，因为这也是你可以采用的另一条规则，即你可以将自己定位为 Logora 平台的维护者。对吧？比如，我们把 Logora Web 应用里的所有功能都以 API 形式提供，然后你们可以随意调用。这是你们做出的承诺，还是说——

英文原文

>> The API business is actually interesting like like you said because you know, that's another rule that you can have where you will see TL maintainer of the Logora platform. Right? Like we will offer everything that we have in the Logora web app as an API and then you guys can use whatever. Is that is that a promise you make or

这不是我们目前承诺的事情。我们确实有可供对接构建的 API，但并没有覆盖产品的全部功能面。

英文原文

>> That's not something we're promising now. We do have APIs that you can build against, but not the entire surface of the product.

这正是令人好奇的地方，这是既定策略还是……显然，有些人会很喜欢它，因为“我不需要学习新工具，直接用现有的工具就行了”。对吧？然后再比如安装 Logora 处理 Word 文档那些东西。除此之外，你知道，我觉得后续还得去维护它。

英文原文

>> That's the curious thing like is that the strategy or and obviously you know, some people will love it because I don't have to learn a new tool. I can just use the the existing tool I have. Right? And then it just like install Logora for example with the word documents stuff. Other than that, you know, I do think like then you have to maintain it.

没错。就是这样，这正是我的顾虑。这也是我们一直尽量不这么做的原因。我们过去一直没有 API。我是说，我们大概是六个月前左右才开始做 API 的。我们抵触了很长时间。主要是因为，是的，我只是不想——我们在产品上的迭代速度太快了，我不想因为必须为已变更的数据模型维护旧版 API 而被拖慢脚步。我的意思是，现在我们规模更大了，情况有所不同。所以现在这是一笔我们可以承担的成本，但当我们只有 20 名工程师时，我们迭代实在太快了，根本做不到这一点。

英文原文

>> Exactly. That's that's so that's been my thing. That's why we've tried not to do it. We haven't we haven't had APIs. I mean we started our APIs like six months ago or something. We fought it for a very long time. Mostly because yeah, I just don't want to we we move so quickly in in the product that I don't want to be like slowed down by having to maintain old API versions for like a data model that we've changed. I mean now we're bigger and things are different. So now it's it's a cost that we can take on, but like when we're 20 engineers it's just like we we move so quickly that we can't do this.

我琢磨过这个问题，对吧？一般而言，所有公司同时拥有平台团队和产品团队都是有好处的，因为产品团队你有一个，但只要平台不过度与产品紧密绑定，你就可以轻松启动第二个。比如最简单的例子就是 Web 应用和移动应用。或者像 Web 应用、移动应用、TCP，然后再接其他各种集成。那样的话，无论如何你的 FD 只是整个平台的又一个客户端而已。

英文原文

>> I have toyed with this, right? Like it's just generally good for all companies to have a platform team and a product team because the product team you have one, but then you can easily spin up a second one as long as the platform is not super tight to product. You know, like a very simple one is web app mobile app. Or like web app mobile app TCP and and then like you know, whatever like other other integrations. Then your FDs just are one more client to the overall platform anyway.

而且我认为，即便是 Agent，你也可以这样思考：有一个类似运行 Agent 的 Harness 团队，然后你甚至可以围绕它构建产品，赋予它各种技能、自动化以及其他任何东西。

英文原文

>> And I think like even the agent you can think about it this way, there's like a Harness team or like a that runs the agent and then like you can even you can build a product around it with like skills and automations and whatever else.

是的，这就非常像著名的亚马逊 AWS 禁令，贝索斯当时规定不再允许后门访问，只能向其他团队提供你们的 API，也就是向客户提供的同款 API。如果你没有，就去建一个。

英文原文

>> Yeah, this this is like very famously the Amazon uh, web services mandate but like Bezos was like no more backdoor access, you only offer other teams your APIs, the same APIs you offer customers. If you don't have it, you build it.

是的。

英文原文

>> Yeah.

然后客户也能用上它。

英文原文

>> And then the customers also get it.

是的。我认为这也正在发生一些变化。现在的很多 API……比如我们在琢磨的一件事就是：Agent 本身能不能成为 API？我知道这听起来很奇怪，但对于其中某些事情，他们真正想做的是触发特定事件的发生。对于其中一部分来说，这完全可以——我并不是说我们正在这么做，但这确实是我脑子里在琢磨的一件事——它能不能直接变成：好，你触发 Agent，让它去做这件事。显然，这比直接调用一个现成的 API 效率要低得多，但这是我正在琢磨的事情，也就是我们的产品中有多少可以仅仅作为 Agent 存在，而不是那些杂七杂八的东西。

英文原文

>> Yeah. I think I think that's also changing a little bit. A lot of um, a lot of APIs now like one thing that we're toying with is like can the agent be the API? I know it sounds weird but like for some of this stuff what they actually want to do is they want to trigger certain things to happen. For some part of this it could it could just be I'm not saying that we're doing this but it's like a thing that I'm toying with in my head is could that just be like okay you trigger the agent and ask it to do this thing. Like obviously it's way less efficient than if you just had like an API that you always hit and it's like you know, but it's something that I'm toying with is like how much of our product can just be the agent versus all the other stuff.

嗯，你知道，无限制的执行、成本和延迟。那就是阻碍……这是唯一的阻碍吗？

英文原文

>> Well, you know, unbounded execution and costs and latency. That's the That's Is that the only holdback?

是的，还有确定性问题，比如在一百次里面可能会有一两次做出我们不希望它做的反常举动。

英文原文

>> Yeah, and then I think deterministic like maybe in you know, 100 out of or in two out of 100 times it's going to do something whack that we didn't want it to do.

产品克制与战略取舍

我们稍后会深入探讨 Harness 和模型微调，但我觉得这里有一个关于克制的产品问题，即你选择不做什么。因为你本来可以把那些全都做了，显然这里存在一个优先级问题。你们现在更倾向于对大多数事情说“不”，只做极少数事情；还是实际上想做更多事情，只需要再招 100 个人？

英文原文

>> We're going to go into like Harness and and model fine-tuning in a bit but I think there's a there's a product question about restraint but what you don't do because you could do all that and obviously there's a prioritization issue like it Are you at more in a point of let's uh, say no to most things and then only do a short number of stuff or no actually we want to do more things and we need just need to hire like 100 people.

[笑声]

英文原文

>> [laughter]

很长一段时间里，我都极度倾向于第一种做法。对所有事情说不，保持工程团队……我想我以前说过这个，在我们今天的一场演示中，在我加入 Liquor 半年之后，我拿出了这张幻灯片，上面是电影《斯巴达 300 勇士》，我说在公司未来的整个历史中，我们只保留 20 名工程师，我们要击败所有规模远大于我们的对手。而现在我们已经有 100 名工程师了，感觉还是远远不够。我本来希望那样，但我认为现实世界中，我们处于激烈的竞争之中，人们想要的东西太多了。如果我像个清教徒一样坚守“不，我们必须保持 20 个人，而且我们只能做某某东西”，那我们就会输掉，客户就会流失到其他地方。所以，我们必须做更多东西。因此，我认为我的职责是去探索如何构建一个系统，让我们能够做很多东西，但实际上只做……在客户感知中我们构建了海量功能，但实际上我们只构建了少数几样东西。这就是我现在对自己工作的思考方式。

英文原文

>> I was I was very strongly the first thing for a very long time. Let's just say no to everything like keep the engine team I I think um, I've said this before but like one of our presentations today like after I've been at Liquor for half a year I I pulled up this slide of um, you know, the 300 Spartans and and and for the movie and I was like we're just going to be 20 engineers for the rest of the company's history and like we're going to beat everyone who's much bigger and we're 100 engineers now and it's just like we're way too small. I wanted it that but I think the in the real world, we're in like fierce competition, people want so much stuff. It's like uh I could be a Puritan and just be like, "No, we have to stay 20 people and we can only build, you know, whatever." Uh but then we just lose. Uh and like people would go elsewhere. So, we have to build more. And so, I think my job is to figure out how do I build a system where we can build a lot, but actually only like where we can the customer feels like we've built so much stuff, but we've actually only built few things. That's That's how I think about my job right now.

是的。也就是原语，对吧？

英文原文

>> Yeah. I mean, primitive is right, you know?

对，没错。我认为这将会……这真的非常强大。

英文原文

>> Yeah, exactly. I think that's going to be That's really powerful.

会测试新产品吗？只是看看，“嘿，这里有没有任何吸引力？”然后你就有点像……你说，“不，我们还是专注于核心业务吧。”

英文原文

>> Do testing new products, just seeing if like, "Hey, there's like any pull here?" And then you just kind of like you're like, "No, let's just focus on our main thing."

噢。我们还没有做过——我的意思是，我们做过某些战略押注，

英文原文

>> Ooh. We haven't made I mean, we've made certain strategic bets,

是的。

英文原文

>> Yeah.

但我们从未尝试过任何我们尚未抱有充分信心的东西。主要是因为我们面临的需求堆积如山，都在喊着“我们想要所有这些东西”。所以对我们来说，去尝试“嘿，也许这东西能行”并没有真正的意义，因为有太多人在大声呼唤其他东西了。因此，我们主要构建那些需求。有时我们也会与客户产生分歧，我们会认为：“实际上，我认为这里的这个东西在战略上非常有意义。”然后我们无论如何都会去尝试构建它。

英文原文

>> but we've not tried anything that we don't have a lot of confidence in already. Mostly because we just have like a mountain of demand as like, "We want all of these things." And so, it doesn't hasn't really made sense for us to try to be like, "Hey, what Maybe this thing could work." cuz like there's so many people screaming for the other things. And so, we mostly uh build those. And then sometimes we'll disagree with clients and we'll be like, "Actually, I think this thing over here will make a lot of sense strategically." And then we'll try to build that anyway.

比如对你们而言在战略上有意义。

英文原文

>> Like strategically for you.

对，没错。

英文原文

>> Yeah, exactly.

那么，这些东西具体指什么？

英文原文

>> So, what are those things?

其中之一实际上就是 Agent。当时其实没有人提出“我希望正在使用的助手具备更多自主性，不需要我吩咐就能做事”这样的要求。根本没有人这样提过。但我们当时就坚信：“这绝对是行业未来的发展方向。这将会赋能……”

英文原文

>> So, one of those things um was uh the agent, actually. So, like no one was actually asking for like, "I want it to be uh I want the assistant that I'm using to have more agency and do things without me asking it." Like no one No one ever But we were just like, "This is definitely where the industry is going. This is going to empower

让你能做很多很酷的事情。”嗯，协同合作是另一个方面，我们觉得：“我们认为整个行业正在走向这样一个方向，也就是你可以在自己工作的系统内部进行协作，也就是 Legora 里面，而不是在邮件和短信这类地方。”所以，是的，确实有这么几个方面。

英文原文

you to do so much cool stuff." Um collaboration was another one where we're like, "We think the industry is going to a place where you can collaborate inside of where you're working, which is Legora, and not in like emails and text messages." So, yeah, there there are there are a few of those.

然后具体是怎样的呢？一旦你们说：“好，是的，我们希望你们大家使用智能体。”我猜测在律所内部，你知道的，要让他们真正学会如何去使用它，需要付出很多。所以，这个时候你们就会类似说：“嘿，FD 会去那边教他们能做些什么。”

英文原文

>> And then how like once you say, "Okay, yeah, we want you guys to use agents." I'm assuming like internally at the law firm, you know, it's going to take a lot for them to actually learn how to do that. And so, that's where you kind of like say, "Hey, FD is going there and teach them what they can do."

主要是工程师。他们会 是的，是的。

英文原文

>> engineers mostly. They'll be Yeah, yeah.

确实是这样。

英文原文

>> That makes sense.

如果你愿意，有人称他们为前线部署法律工程师，但本质上他们就是做这个的。嗯，然后他们进驻 是的，并帮助并协助他们围绕智能体的各项能力来构建其工作流。

英文原文

>> some people call them forward deployed legal engineers, if you want, but that's essentially what they are. Um then they go in Yeah, and help and help them structure their workflows around uh the capabilities of the agent.

模型微调与工程框架

你们之前说过，我想 Max 提过这点，早先具体来说你们并没有在微调自己的模型。现在依然是这种情况吗？

英文原文

>> You guys said previously, I think Max said this, but that early on specifically you guys weren't fine-tuning your own models. Is that still the case?

是的，到目前为止我们还没有微调过任何模型。我认为在过去的几个月里，开源权重模型已经变得能力极其强大，因此针对特定任务我们确实会做微调，特别是当我们认为在前沿模型以对应成本和延迟所能实现的效果，与我们认为后训练模型所能实现的效果之间存在巨大差距时。法律检索就是一个绝佳例子。它就像一个 你必须做 这是一个扇出式检索问题，你可以打造一个非常高效的模型，以极快的速度、极低的成本和极低延迟来执行这些查询。

英文原文

>> Yeah, so we haven't fine-tuned any models up until now. I think now in the past few months open weight models have become extremely capable and so we will do it for certain tasks where we think there's a huge delta between what a frontier model can do at the cost and latency versus what we think post-trained model can do. Legal research is a great example. It's like a you have to make it's a it's a fan out retrieval problem and you can make a really effective model to do those queries super quickly and super cheaply at low latency.

所以具体而言，你们对它们进行后训练是为了让它们更擅长于……

英文原文

>> So specifically you're post-training them to be better at

非常具体、有边界的任务。对，正是这样。

英文原文

>> very specific bounded tasks. Yeah, exactly.

明白了。

英文原文

>> Okay.

我们并不是试图去训练我们自己的 Fable。

英文原文

>> We're not trying to like train our own Fable.

是的。

英文原文

>> Yeah.

让它去充当主力智能体。我不认为那是……我们很可能在这方面竞争不过 Anthropic，你知道的。

英文原文

>> To like be the main agent. I I don't think that's like a We won't win that, you know, versus Anthropic most likely.

仍然有大量的，比如说搜索与检索研究，或者说 RAG 研究需要去做，这或许没有那么光鲜。有时它甚至实际上都不属于模型训练，而纯粹是架构和管线层面的工作。你们和 Turbo Buffer 做过一个案例研究，不过，你知道，还有像你们是否在做后期交互（late interaction）……

英文原文

>> There's still a lot of let's say search and retrieval research or rag research to do that is like maybe less glamorous. It's not actually even model training sometimes. It's just like architecture and pipelines. You you you did a case study with Turbo Buffer, but like, you know, there's also like do you do late interaction of

是的，确实如此。你们确实做这个。是的，我们实际上没做那个，但你说的完全正确，这就是……团队里有个人叫 Sean，他就是特别关注搜索这一块，他基本上经常跑来跟我说，类似：“噢，我们应该做这些事情。”然后这是新的嵌入模型，比如我们可以做这种后期交互。他实际上跟 Turbo 团队沟通，让他们开发了多向量支持，以便我们能够实现这类事情。

但不，在模型之前，甚至老实讲在整个运行框架（harness）里，普遍存在着大量的低垂果实。比如如果你只是想为我们的客户打造更好的输出，在训练模型之前，我们有太多的事情可以做，并且能带来立竿见影的效果。

英文原文

>> Yes, absolutely. You do that. Yeah, we actually don't do that, but it's you're completely right that this is So there's one guy on the team, Sean, who's uh he just really cares about search and he he basically comes to me all the time and he's like, "Oh, we should do these things." And then this is new embedding models and like we can do this late and late interaction. He actually talked to Turbo teams to get them to build multi-vector so that we could do these types of things.

But no, there's loads of low-hanging fruit generally even before models and even in the harness honestly. Like if you just want to create a better output for our clients, there's like so much stuff that we can do that has immediate impact before training models.

比如针对法律行业，什么样的框架才算好框架呢？我猜你们是在构建自己的定制框架。

英文原文

>> Like what makes a good harness for like the law industry I guess? Like you guys are making I'm assuming like your own custom harness.

是的，完全没错。是的。我们编辑文件的方式很重要；我们进行分析的方式很重要；做法律检索的方式全部内嵌在其中。所以这就好比，你可能会遇到一个冒出来的问题或者什么情况，然后它懂得如何去进行法律检索，从而找出答案是什么、先例是什么。这些都非常重要。当然，围绕它的所有治理机制也同样包含在内。

英文原文

>> Yeah, absolutely. Yeah. The way that we edit files is important. The way that we do an analysis is important. The way you do legal research is all baked into it. So it's like you might have a problem that comes up or you something, then it knows how to go and do the legal research to figure out what the answer is, what the precedent is. That's all very important. And then of course, there's all the governance stuff around it as well.

是的，看起来搜索可能是你们真正深度定制的最重要的工具，因为……

英文原文

>> Yeah, it seems like the search is probably like would you say the most important tool that you guys are really customizing cuz

它是……

英文原文

>> It's

如果你无法调出正确的信息……

英文原文

>> if you can't pull in the right information

那你就完蛋了。你就完蛋了。这……我认为它是、它是……嗯……

英文原文

>> then you're screwed. You're screwed. It's uh I think it's it's it's um

你们正在检索海量的文档。

英文原文

>> You guys are searching like a lot of documents.

检索、起草与复杂合同

是的，可能是搜索、分析，以及起草。因为在 Word 里起草大篇幅合同，其实也绝非易事。

英文原文

>> Yeah, it's search, it's analysis, and it's drafting, probably. Cuz like drafting word drafting huge contracts is actually also not super trivial.

比如具体哪方面不简单呢？

英文原文

>> Like what about it is not trivial?

比如假设你要从零起草一份庞大的合同，那你必须去寻找先例，比如我们之前是怎么处理这个的？我们是否需要这种类型的条款？

英文原文

>> have to let's say you're drafting a huge contract from scratch, then you have to find precedents of like how do we do this previously? Do we need this type of clause or not?

律所？他们之前是怎么做的？

英文原文

>> law firm? How did they do it previously?

比如说我就是这家律所，我正准备起草一份庞大的随便什么合同。保密协议（NDA）非常简单，但存在着复杂得多的合同，这时你首先需要找到合适的先例，因为律师根本不会就这么坐下来然后开始打字。他们会弄清楚在类似的交易中使用了哪些条款，而在搜索中如你所知，“类似”可以代表很多不同的含义。它可以是相同行业、相同类型的公司，或者相同其他什么属性。嗯，所以你需要甄别出正确的先例。你可能会拥有多份文档，然后试图理清应该从哪份文档中抽取哪部分内容。而这实际上绝非易事。所以，它本质上也是一个搜索问题。

英文原文

>> so I I am the law firm and I'm about to draft a huge whatever. NDA is very simple, but like there's there are contracts that are way more complex, and then you need to find the right precedent first cuz like lawyers just don't don't I sit and like start typing stuff. They they figure out what clauses were used in similar deals, and similar can mean a lot of different things as I'm sure you know in search. It could be like same industry, same types of companies, same whatever. Um and so you need to identify the right precedent. You might have like multiple documents that you then try to figure out which part should I take from what document. And that's actually non-trivial. So, it's essentially also a search problem.

模型起草表现与幻觉校验

哪个模型最擅长起草？

英文原文

>> Which model's best at drafting?

[笑声]

英文原文

>> [laughter]

这么说吧，我昨天部分提到过这一点。某些受偏爱的模型确实话特别多，我相信你们都知道这点。呃，它们非常啰嗦。

英文原文

>> So, I'll say I told this partly this yesterday. Definitely favorable models are extremely chatty. I'm sure you guys have known this. Uh they're very verbose.

这可不好。

英文原文

>> not good.

这通常不太好。是的。

英文原文

>> That's generally not good. No.

我注意到 Astra 在写作方面相当出色。

英文原文

>> I've noticed Astra's pretty pretty good at writing.

Astra 是的，写作相当不错，要精炼得多。嗯……

英文原文

>> Astra's yeah, pretty good at writing. Much more more concise. Um

你们是否运行了比如一百个不同的提示词来……是的，来确认这一点？

英文原文

>> Are you running like a hundred different prompts to like Yeah. Be sure.

我们能 是的，我们能做某些处理，但我认为啰嗦冗长这种倾向实际上让我们有点难办。所以，这里我认为，你知道，走着瞧吧。但是 Anthropic 的下一代模型大概率在这方面会有很大改善。Astra 实际上非常精炼。

英文原文

>> We can Yeah, we can we can do we can do certain things, but I think the chattiness verbosity is actually a little bit difficult for us to do. So, this is I think you know, let's see. But, probably next model from Anthropic is going to be much better at this. Astra is actually very concise.

对，对。我猜他们全都和你们建了 Slack 频道，你们也在向他们反馈这些，对吧？

英文原文

>> Yeah, yeah. And I I assume they they all have a Slack channel with you and you're telling them this, right?

正是。是的，我们会告诉……

英文原文

>> exactly. Yeah, we tell

他们以前没接触过这类情况。

英文原文

>> They're not used to them.

不，不，不，他们清楚这个。

英文原文

>> No, no, no, they know this.

另外一件我非常关心的事情就是幻觉。它们会捏造内容吗？

英文原文

>> Then the other thing that I really care about is hallucination. Do they make stuff up?

所以，呃，我们有一套引用系统，基本上会核对它们所写的内容是否有链接对应。这样用户就拥有一个可以点击的链接，点击后会用边框在 PDF 或类似文件中把信息提取自哪里的具体位置高亮显示出来。我们能在一定程度上确保它所说的内容在某处始终是可引证的，甚至是往下追溯多层。所以，打个比方，如果智能体根据某个表格审查写了某些内容，你知道，它可以引用该表格视图中的具体单元格，或者表格视图中的那一行。因此你实际上可以往下点击多层，去找到，你知道，去一路往上验证它。

英文原文

>> So, uh we have a system of a citation system where we basically check what they're writing has a link. So, the user has a link they can click and it highlights with a bounding box like in the PDF or whatever. Where where this information was taken from. And we can sort of make sure that what it says always is citable somewhere. Even multiple layers down. So, like if the agent writes something based on a tabular review, you know, it can cite the cell in the tabular view or the row in the tabular view. So, you can actually like click multiple layers down and find, you know, sort of validate it all the way up.

是的。

英文原文

>> Yeah.

你有没有发现律师们究竟是否在核实这些引用的准确性？

英文原文

>> Are you finding that lawyers are like verifying that these citations are correct or not?

起初……起初有一部分人会核实。

英文原文

>> Initially Initially Some of them were.

是的，确实如此。都是同样的情况。

英文原文

>> Yeah, exactly. It's the same thing.

你真的……

英文原文

>> Are you actually

大家应该……比如幻觉出指向最高法院的链接，他们根本不在乎。

英文原文

>> People should like hallucinated links to Supreme Court. They don't care.

这种情况确实会发生。是的，确实如此。那种事是会有的。[笑声]

是的，我觉得这就像写代码一样，你知道的，有的工程师就是根本不在乎，直接让智能体跑；而有的人则会坐下来逐字逐句看每个逗号。分布情况完全一样。

英文原文

>> That happens. Yeah, exactly. That happens. [laughter]

Yeah, I I think it's like coding, you know, you have engineers that just don't give a and just like send agents. And you have people that like sit and read every comma. Same distribution.

数量如此庞大，单靠一个人怎么可能看得过来，对吧？

英文原文

>> like so much that like how can one human do that, right?

是的，但我认为他们大概就像我们在代码审查中知道该看什么一样。

英文原文

>> Yeah, but I think they probably just like We know in code review what to look for.

是的。

英文原文

>> Yeah.

比如系统是什么等等。我认为他们也清楚，好，这就是关键所在，这些是危险信号，这才是重点内容，让我重点复核一下这几个地方就行。而且这些听起来挺合情合理，所以大概率没什么问题。

英文原文

>> What is the system or whatever. I think they also know, okay, this is the the thing. These are the red like this is the thing that's important. Let me just like double check these things. And like these sound reasonable, so it's probably fine.

好的。挺有意思。

英文原文

>> Okay. Interesting.

就像如果你从初级律师那里收到一份文件，你做的也是同样的事情，对吧？就像：这个地方我得复查一下，而这部分应该没什么问题。

英文原文

>> Just like if you if you receive something from a junior lawyer, it's the same thing you do, right? It's like I should double check this. And like this probably is fine.

长文档推理与下一代模型期待

还有没有其他维度的，姑且称之为法律推理，或者说在法律领域的前沿，有哪些是模型目前做得还不够好，而你们希望下一代模型能够具备的？

英文原文

>> Any other elements of let's call it legal reasoning or what's the the the legal frontier that models are not good enough yet that you want from the next generation of models?

目前我并没有想到特别具体的点。我认为长文档推理目前依然很困难。

英文原文

>> Nothing I can think of specifically right now. I think Long document reasoning is still difficult.

嗯，长到比如百万 token 的上下文就像……

英文原文

>> Um long like million token token context is like

是的，正是如此。你手头有一份庞大的合同。

英文原文

>> Yeah, exactly. You have a huge contract.

确实。

英文原文

>> Yeah.

嗯，而且比如你要通篇进行修改。这对当今的模型来说依然是一件棘手的事情，即使……并且比如要记住你修改了所有这些

英文原文

>> Um and like you're making edits across it. It's a difficult thing to do for a model today. Even. And like remember that you changed all these

这里的东西，比如定义改了，然后顺序改了，还有其他诸如此类的情况。

英文原文

things up here and like the definitions changed and that you the ordering changed and whatever else.

那么，最简单的办法就是单次提示完成。你把所有内容喂进去，然后一次性输出。

英文原文

>> Um so, the simplest would be one shot. You put feed everything in and then one shot output.

是的。

英文原文

>> Yes.

但你们并不这么做。你们把它拆分成块，然后编排这些修改。

英文原文

>> You don't do that. You break it up into chunks and then you orchestrate edits.

是的，没错。正是这样。这也是在这个行业里很有趣、参与其中觉得好玩的事情之一，对吧？因为我的意思是，我们做过，你知道的，我们做过 MapReduce，后来又变了，那是一次重大转变，而且经历了多次类似的大变动。随着模型的改变，我们围绕它们构建的脚手架也在不断调整。

英文原文

>> Yeah, exactly. Exactly. And this is like one of those things that are funny like that are fun to be part of in this industry, right? Cuz our word that I mean was like we've done, you know, we've done map reduce and we've changed and that was one big one and that was like one big one in multiple times. It's like we keep changing the the scaffolding that we build around as models change.

是的。我……我想说我最近……我显然因为自己的工作做了很多类似新闻报道的事情，而我最近已经从 MapReduce 转向直接把所有东西都丢进上下文窗口里了。

英文原文

>> Yes. I I I would say I I um I recently I do a lot of like news reporting obviously because of what I do and I've recently shifted from map reduce to throw everything into context.

而且效果更好。

英文原文

>> And it's it's better.

确实。但这也……情况不一样。我是说，我们在搜索上遇到过这个问题：你在庞大的合同中检索某些内容，你可以用 MapReduce 来处理，但如果是在做任何形式的总结，它就可能会丢失块与块之间的一些重要细微差别。

英文原文

Exactly. But it's also it's different. I mean we had this problem with search where it's like you're searching for certain things in huge contracts and you can map produce it but then it might if you're summarizing any sort of thing then it might lose some important nuance

对。

英文原文

>> Yep.

尤其是如果你只是把各个单独的部分汇总起来的话。所以我们……这一直是个问题，就变成我们可以做两种类型的搜索。我们也可以做另一种，也就是在中间的不同步骤中需要有推理。我想举一个很贴切的例子。是的，有些文件里，他们在两页左右的篇幅里详细讨论某件事，单看每一个单独的文本块，未必与搜索词或他们试图搞清楚的事情相关，但组合在一起看时，它就变得极其相关。所以，是的，这些就是所谓的部分难题，在这种情况下……

英文原文

>> in between the chunks. Essentially, if you just summarize the individual parts up. And so we've this has been a problem where it's like we can do two types of search. We can also do the other one where it's like um you need to have reasoning in in the different steps there. in between uh so I'm trying to think of a really good example here. Yeah, there's there's some documents where they talk about something at length over like let's say two pages or whatever and then each individual chunk is not necessarily relevant to the search term or what they're trying to figure out but together it's extremely relevant. And so yeah, these are some of the you know problems where it's

那肯定很难，就像你之前说的，要真正让模型在这方面表现更好，因为你们甚至都不能去看那些合同，对吧？因为那是真实的合同。那么你们会向他们提出请求吗，比如：“嘿，你们能制作一些和你们真实合同类似的样本合同吗？”

英文原文

>> And that must be hard like you said earlier to actually make the models better at that cuz you can't even look at the contracts, right? Cuz they're real contracts. So do you guys ever ask them for like, "Hey, can you make example contracts that are similar to your real ones?"

会的，是的，我们有反馈机制，我们可以收到他们的反馈，如果他们愿意，或者他们只是说：“嘿，这对我们超级重要，但现在效果不好。”那我们就会说：“好，太棒了。你们能对文档以及诸如此类的问题进行脱敏处理吗？只要我们能复现它，我们就能修复它。”

英文原文

>> Yeah, yeah, so we have a we have feedback and we can receive feedback from them and if they're willing to or they're just like, "Hey, this is like super important to us and it's not working well." Then we're like, "Okay, great. Can you like anonymize the documents and like the problem then whatever and we'll we if we can recreate it then we can fix it.

是的。我是说，这种事情的一种解决办法就是你们去买下一家律所。

英文原文

>> Yeah. I mean so one version of this is you buy a law firm.

对。

英文原文

>> Yep.

很多人都在考虑这么做。

英文原文

>> Which ton of people are considering.

是的。但是……关于他们的那些数据到底算不算他们的数据，还是算客户的数据，这里面有一些问题。

英文原文

>> Yes. But uh there's some we there's some stuff there on whether uh their data is actually their data or if it's the client's data.

嗯。

英文原文

>> Hm.

所以如果客户都没了，那或许还可以。其实我也不知道。

英文原文

>> So if the clients are all gone then maybe it's fine. I actually don't know.

有意思。

英文原文

>> Interesting.

是的。但这是……是的，这是获取大量数据的一种方式。

英文原文

>> Yeah. But it's uh yes. It's it's a way to get a lot of data.

是啊，大家还说谷歌要收购精神航空（Spirit Airlines）呢。

英文原文

>> Yeah, yeah, people are like Google's buying Spirit Airlines.

对吧？

英文原文

>> Right?

是啊，他们买了……对，花了大概 1000 万到 700 万美元之类的。对，对。

英文原文

>> Yeah, they bought Yeah, they bought the Yeah, yeah, for like 10 to 7 million or whatever. Yeah, yeah,

就像你知道的，如果你开公司倒闭了，你还有 Slack 记录，你可以把 Slack 卖掉。我觉得好像有一家 YC 孵化的公司专门收购你倒闭的公司。

英文原文

>> Like you know, if you have a failed company, you still have your Slack, you can sell the Slack. Like, I think there's like there's like a YC company that like buys your failed company.

是的，确实有。对，对，对，对。

英文原文

>> Yes, there is. Yeah, yeah, yeah, yeah.

[笑声]

英文原文

>> [laughter]

是啊，因为大家就像是在想：“哦，我只想要真实的人在干活的数据。”

英文原文

>> Yeah, cuz like it's just like, "Oh, I just want like real people doing work."

确实。

英文原文

>> Exactly.

然后就能产出很酷的东西。

英文原文

>> And then there's just like cool stuff out.

没错。购买律所数据的问题在于……好吧，你能拿到所有的输入数据，真实的输入，这很好。但输出就相对难办一些，因为你拿到的只是他们的最终成果，很难去弄清楚……你基本上得逆向推导他们是如何得出那个结果的。因此，在我们的产品中，当我们拿到真实用户的真实追踪轨迹时，其价值极其巨大。但如果只拿到这么一份输出，我们既不知道它是好是坏，也不知道它是怎么生成的，这就有点棘手了。

英文原文

>> Exactly. The problem with buying uh law firm data is do Well, you get all the inputs, with the real inputs, which is great. Uh the outputs are a little bit more difficult cuz you just get what their end result was, and it's difficult uh to figure out if like the way you have to basically like uh reverse engineer how they got there. So, in our product, when we get real traces from real users, this is extremely high value. But, if we just got this output, and we don't know if it's good or bad or how it was created, it's a little bit difficult.

嵌入模型与行业预测

我们很快就要收尾了。最后我想向大家指出的也是，在嵌入模型方面其实有很大的收益空间。如果你的工作负载严重依赖 RAG，而你只是看着 MTEB 榜单然后挑排名靠前的模型，那你就错失了很大一部分潜在优势。你应该去看看 Voyage 或者一些定制的嵌入模型。甚至听听这个。

英文原文

>> We're going to wrap up soon. Uh the last thing I'll point people to is also there's a lot of alpha lesson embeddings. If you're if you're rag-heavy work workload, uh you're you're just looking at MT B and you're taking like top model, uh you're leaving a lot on the table. You should look at like uh Voyage or like some of the custom embeddings. You even listen to this.

我们……是的……

英文原文

>> We've Yeah, uh

是的，所以基本上不要直接用默认的嵌入。即使是 OpenAI 的嵌入，OpenAI 的嵌入也无所谓。定制你的嵌入，对吧？

英文原文

>> Yeah, so basically don't take default embeddings. Even like OpenAI embeddings, OpenAI embeddings, doesn't matter. Customize your embeddings, right?

对。

英文原文

>> Yeah.

是的。

英文原文

>> Yes.

你会支持这一观点的。

英文原文

>> You you will you will support that.

是的，我认为这没错。我觉得是对的。唯一的一点是需要确保你支持多语言。对我们来说，我们要服务非常多的语种，所以这相对来说更成问题一些。

英文原文

>> Yes, I think that's right. I think it's right. With the only thing is uh need to make sure that you're multi language. So, for us, we we serve so many languages, so that's a little bit more of a problem.

这方面的王者。

英文原文

>> king of this.

是的，没错。

英文原文

>> Yeah, yeah, exactly.

那个 Mistral 对这方面有帮助吗？

英文原文

>> Uh Mr. Owl helpful there?

呃，没有。

英文原文

>> Um no.

[笑声]

英文原文

>> [laughter]

直到……好吧，直到目前为止，他们刚融了一大笔钱，对吧？所以，也许……

英文原文

>> Until Well, until now, they just raised a ton of money, right? So, maybe

是啊。

英文原文

>> Yeah.

我是说，你们最好……

英文原文

>> I mean, you guys better

一年前融了一大笔钱。

英文原文

>> raised a ton of money a year ago.

你们会成为你们的支持者，比如……

英文原文

>> You're going to be your champions, like

我觉得我们需要更多欧洲本土出来的模型公司。是的。

英文原文

>> I think we need way more model companies out of the out of Europe. Yes.

我只是想问问你，对未来一年的法律行业有什么预测吗？

英文原文

>> I just wanted to ask you like what predictions you have maybe like over the next year for like the law industry.

呃，其实我们早些时候聊过一点这个。我把它改成你的第二个问题，那就是：两年后，律师们是在 Legora 内部开展工作，还是会使用其他工具比如 Claude 或 ChatGPT，然后调用 Legora？就像 API 调用那样。

英文原文

>> Uh actually we we asked a little bit of that earlier. I'll modify it to your second question, which is uh in 2 years will lawyers do their work inside Legora, will will they use uh other tools like Claude or ChatGPT and call Legora? Like, you know, the API call.

对，对，对，对，对。

英文原文

>> Yeah, yeah, yeah, yeah, yeah.

我觉得两者都有。这真的视情况而定。我认为如果你是一家大企业，你可能会用 Claude 协同办公，然后你……我觉得这有点像我把 Linear 用于 Claude，但我依然会去打开 Linear。我认为企业会这么做。而我认为律所则会直接使用 Legora。因为那样更简单，也更好。

英文原文

>> I think both. It really depends. I think if you're an a big enterprise, you might have Claude co-work, and then you'll it's I think it's similar to like how I use Linear for Claude, but I still go to Linear. I think enterprises will do that. I think law firms will just use Legora. Because that's simpler and better.

是的。

英文原文

>> Yeah.

这有点让我想起医疗领域的 OpenEvidence。你们基本上就是法律领域的 OpenEvidence。

英文原文

>> It kind of reminds me of like open evidence for like medicine. Like you're you're basically that for law.

是的。

英文原文

>> Yeah.

是的，没错。

英文原文

>> Yeah, exactly.

在代码领域，大家都已经到了各大实验室会因为你与他们竞争而切断供应的地步。这种情况已经发生过两次了。

英文原文

>> In coding is all into a point where people the labs will cut you off for competing with them. And this has happened twice.

是的。

英文原文

>> Yeah.

你觉得这种事会发生在法律领域吗？

英文原文

>> Do you Do you think that will happen to law?

不，我不这么认为。我觉得情况就像……你知道他们正在向他们发布法律服务。

英文原文

>> No, I don't think so. I think it's like you know they are releasing legal services to them.

是的，对，对，确实是。我不是……我是说我们显然经常跟他们交流，但这也……这就像“嘿，你们打算……”

英文原文

>> Yeah, yeah, yeah, yeah, they are. I don't I mean we obviously talk to them a lot, but it's also it's like hey are you going to

[笑声]

英文原文

>> [laughter]

是啊，你们是销售服务……

英文原文

>> Yeah, you're sales services

不。

英文原文

>> No.

就像他们做不了决定。

英文原文

>> Like they don't make the call.

没错，那不是他们的……

英文原文

>> Exactly, that's not their

[笑声]

英文原文

>> [laughter]

对，而且他们很多时候甚至都不知道到底在发生什么。

英文原文

>> Right, and they don't even know like often what's actually going on.

我不认为……我不认为情况是那样。他们在知识工作或者广义的企业知识工作中需要覆盖的内容太多了，以至于法律必须成为其中的一部分，就像数据分析、销售、营销和财务也必须是其中一部分一样。所以他们必须在这些方面做出不错的东西。但他们绝不会像我们做得这么深入。而且我认为，如果 Claude 接下一些小客户、小企业客户，这些客户通过 Claude 处理一部分法律事务，那很好，这没问题。但如果你要做的是一笔 4 亿美元的并购交易，你肯定会想要 Legora。

英文原文

>> I don't think I don't think that's the case. They have so much stuff to capture in knowledge work or in in sort of enterprises knowledge work generally that like legal has to be part of that just like data analytics has to and sales and marketing and finance. So like they have to do things that are good there. But they're never going to go as deep as we are and I think if if Claude takes small clients, small you know, corporate clients and they do some of their legal via Claude, like that's that's great, that's fine. But if you're doing a you know, $400 million M&A transaction you're going to want Legora.

法学院合作与交叉人才培养

我有一个算是不太相关的问题，但我很好奇，你们会走进法学院，让他们在课堂上教授 Legora 吗？这具体是怎么操作的？你们有在考虑这个吗？

英文原文

>> I had kind of an unrelated question, but I'm curious are you guys going to law schools and kind of like getting them to like teach Legora in classes? Like how does that work? Are you guys thinking about that?

这绝对是一个我们……我认为我们已经启动了的项目。我其实不知道目前的具体进展，但我们确实希望法学院的学生能学会使用 Legora。

英文原文

>> it's definitely a program that we are I think we've started. I actually don't know exactly where we're at, but we do want people in law school to learn to use Legora.

比如你有没有看到法学院的课程体系发生很大变化，还是其实没有？

英文原文

>> Like are you seeing that like law schools are changing their curriculums a lot or not really?

我不知道。我……我其实不太清楚。

英文原文

>> I don't know. I I I I wouldn't know actually.

很有意思。

英文原文

>> Interesting.

我其实会换个方式做 [清嗓子]。你们应该为计算机专业的学生、为工程师设立法学院奖学金。

英文原文

>> I would actually do [clears throat] it a different way. You should do scholarships for for law school for CS majors, for engineers.

他们会资助你的……

英文原文

>> They will cover your

计算机专业

英文原文

>> CS

是的。

英文原文

>> Yes.

我们送你去读法学院。

英文原文

>> We'll put you in law school.

计算机专业加上一些软件开发经验。

英文原文

>> CS plus some software experience.

对，我们送你去读法学院。

英文原文

>> Yeah, we'll put you in law school.

是啊。

英文原文

>> Yeah.

费用我们全包。

英文原文

>> We'll cover it.

或者办那种硕士加法律博士（JD）的双学位项目？

英文原文

>> Or you do like the MA like masters plus JD programs?

是的，但他们不……

英文原文

>> Yeah, but they don't

在乎硕士学位。

英文原文

care about the masters.

[笑声]

英文原文

>> [laughter]

不，我是指，我指的是因为很多人本科不是读计算机的，所以他们就会去读……

英文原文

>> No, I mean I'm meaning like cuz there's a lot of people who didn't do CS undergrad, so then they do like

读硕士只是为了有点像……

英文原文

>> masters just to kind of like get

跳出……

英文原文

>> Step out of

对，对。我，我不知道。我没那么常见到这种情况。是的，有人会读 JD MBA，这确实有。但 JD CS 我还没见过。是的，也有可能。是的，是的。好，总的来说，你看，进展非常令人瞩目。我是说，祝贺你们取得的所有成功。你们接下来还会继续保持这个节奏吗？比如……

英文原文

>> Yeah, yeah. I I don't know. I don't see that as often. Yeah, people do JD MBA. That's a thing. But JD CS I haven't seen it. Yeah, happens. Yeah, yeah. Okay, look overall you know super impressive progress. I mean congrats on all your success. Like are you still going to keep up the pace? Like

扩张与招聘

对，你知道，我想你们正在旧金山、纽约招聘。

英文原文

yeah, you know I think you're hiring in SF New York.

瑞典，对，工程岗位我们在纽约、斯德哥尔摩和伦敦招聘。对，所有其他岗位我们都在到处招人。

英文原文

Sweden yeah engineering we're hiring in New York Stockholm and London. Yeah, and all other roles we're hiring everywhere.

你们现在在纽约有多少人？

英文原文

How many people do you have in New York right now?

100人。

英文原文

100

好的，那你们打算在纽约招多少人？

英文原文

Okay, and you guys are planning on how many in New York?

在纽约？

英文原文

In New York?

对，就大体上来说。就说大体上、总体而言吧。

英文原文

Yeah, just like generally. Let's just say yeah generally in general.

很多。

英文原文

A lot.

尽可能多招。

英文原文

As many as you can.

对，挺进取的。

英文原文

Yeah, it's it's aggressive.

[笑声]

英文原文

>> [laughter]

法律知识工作的未来

对，我是说这就是整个公司法领域嘛，对吧，就全美而言……所以，对，我的意思是我很期待看到你们收购一家律所，然后，你知道，成为……对。我认为，你知道，每个行业，特别是顶尖的，比如医生、律师、程序员，你知道，他们都得有两到三个完全主导该领域的 Agent 实验室。所以你们拥有一个巨大的市场。

英文原文

>> Yeah, I mean that's the whole corporate law right as far as like America's so like yeah, I mean I'm looking forward to you buying a law firm and I you know becoming yeah. I think like you know every industry especially like the top like doctors, lawyers, coders, you know they all have to have like you know two or three agent labs that like just completely dominate the space. So you guys have like a great market.

对，而且我认为随着法律发生改变，以及再次变得能让更多人触达，这个市场只会变得大得多。

英文原文

Yeah, and I think it's only going to get much bigger as like law changes and again becomes more available to more people.

对，另外我觉得，我们时间已经严重超了，但是你们所解决的问题非常属于最前沿的知识工作。

英文原文

Yeah, also I think like we're already like super over but like the problems that you work on are very like the frontier knowledge work.

是的。

英文原文

Yes.

对，它们很……除非你真的在做科学、Navier-Stokes 方程以及科学研究，除此之外，大部分工作基本上都会汇聚到法律上。对。

英文原文

Yeah, they're okay like unless you're literally like doing like science never your Stokes and science other than that like most work is like basically converges to legal. Yeah.

[笑声]

英文原文

>> [laughter]

对，到头来全都是法律。招聘、财务，所有的事，就像我必须咨询我的律师一样，所以……

英文原文

>> Yeah, it's all legal in the end. Hiring, finance, everything like I just have to consult my lawyer so.

对，我知道，我知道，这也是为什么我觉得这是一件非常令人兴奋的研究方向，因为现代社会在某种程度上就是运转在法律之上的。

英文原文

Yeah, I know I know which is why I think it's a really exciting thing to work on cuz like modern society runs in some way on legal.

对，对。太酷了。好了，到这里就结束了。谢谢你。

英文原文

Yeah, yeah. Cool. All right, that's it. Thank you.

谢谢你。

英文原文

Thank you.

多谢。

英文原文

Thanks.

[音乐]

英文原文

>> [music]