An Organizational Second Brain: Building an AI That Learns From Experts

组织化的“第二大脑”：构建一个能向专家学习的 AI

By Shaurya Sengar, Jason Nawrocki, Jay Shah, Prashant Kommireddi 作者：Shaurya Sengar, Jason Nawrocki, Jay Shah, Prashant Kommireddi

We’ve built an AI agent that acts as a secondary expert for a given domain, making deep specialist knowledge readily available and preserved for anyone in an organization to access, share, and build upon. This is not a typical domain-specific agent. Its novelty comes from integrating two layers: A structured, auditable knowledge architecture separates what the agent knows from how it reasons. A self-improvement loop then compiles expert feedback into verified, regression-tested updates without model retraining. Together, these two layers turn one-off expert corrections into permanent, compounding institutional memory, and the pattern is designed to generalize to other domains governed by retrievable text rather than model weights. 我们构建了一个 AI 智能体，它能充当特定领域的“第二专家”，使深度的专业知识变得触手可及，并得以保存，供组织内的任何人访问、共享和在此基础上进行开发。这并非一个典型的领域专用智能体。其创新之处在于集成了两个层面：一种结构化、可审计的知识架构，将智能体“知道什么”与“如何推理”分离开来；以及一个自我改进循环，它将专家的反馈汇编成经过验证和回归测试的更新，而无需重新训练模型。这两个层面共同作用，将一次性的专家修正转化为永久且不断积累的组织记忆，且该模式旨在推广到其他由可检索文本而非模型权重所驱动的领域。

This system is saving domain subject matter experts (SME)s at Meta substantial time, allowing them to focus more on the work where their knowledge matters most. 该系统正在为 Meta 的领域主题专家（SME）节省大量时间，使他们能够更专注于那些最需要其专业知识的工作。

Many large organizations have the same problem when it comes to specialist knowledge. While some of it is written down in the form of models, playbooks, checklists, and frameworks, the most valuable specialist knowledge lives in people’s heads and rarely gets captured anywhere durable. In compliance domains, for example, the same types of questions can arise across hundreds of product reviews, expert assessments take days of manual research, and inconsistency between assessments creates real organizational risk. It’s not uncommon for experts to spend more time answering routine questions than on genuinely novel and ambiguous work where their judgment matters most. 许多大型组织在处理专业知识时都面临同样的问题。虽然部分知识以模型、手册、清单和框架的形式记录了下来，但最有价值的专业知识往往存在于人们的脑海中，很少被持久地记录下来。以合规领域为例，同样类型的问题可能出现在数百个产品审查中，专家评估需要数天的手动研究，而评估之间的不一致会带来真正的组织风险。专家花费在回答常规问题上的时间往往多于处理那些真正需要其判断力的新颖且模糊的工作，这种情况并不罕见。

We need systems that can capture how an organization’s experts reason and make that knowledge available to everyone who needs it, so that expertise is easier to share, build on, and preserve. We set about solving this challenge by codifying institutional intelligence into an AI agent for a specific compliance domain. The agent combines a knowledge system that acts as the organization’s “second brain,” a reasoning layer that mirrors how domain experts actually think, and an automated improvement pipeline that compounds expert effort permanently. The patterns generalize to any enterprise domain with deep specialist knowledge, whether that is finance, security, or engineering. 我们需要能够捕捉组织专家推理过程的系统，并将这些知识提供给所有需要的人，从而使专业知识更易于共享、构建和保存。我们着手通过将组织智慧编码到一个特定合规领域的 AI 智能体中来解决这一挑战。该智能体结合了一个充当组织“第二大脑”的知识系统、一个模拟领域专家实际思维方式的推理层，以及一个能永久积累专家工作成果的自动化改进流水线。这些模式可以推广到任何拥有深厚专业知识的企业领域，无论是金融、安全还是工程领域。

The Architecture at a Glance

架构概览

Off-the-shelf LLMs provide a strong foundation, but they often need deeper institutional context to be fully effective in specialist domains. Without that grounding, a general purpose model has limited value given it will not be able to distinguish between what an organization could do (a summary of general information) and what it should consider doing (based on historic positions, company direction, business context, etc.). In high-stakes domains, closing this gap requires supplying the model with the organization’s own knowledge and priorities so its analysis reflects how the organization actually reasons. 现成的 LLM（大语言模型）提供了坚实的基础，但在专业领域中，它们往往需要更深层的组织背景才能充分发挥作用。如果没有这种基础，通用模型价值有限，因为它无法区分组织“可以做什么”（一般信息的总结）和“应该考虑做什么”（基于历史立场、公司方向、业务背景等）。在高风险领域，要弥合这一差距，需要为模型提供组织自身的知识和优先级，以便其分析能够反映组织实际的推理方式。

The system we’ve designed has four layers, each solving a distinct problem: These layers depend on each other. The knowledge system’s file structure makes automated editing possible. The reasoning layer’s explicit procedures make failure attribution tractable. The evaluation framework gates every change. And the improvement loop feeds back into both knowledge and reasoning. Remove any one layer and the others degrade. 我们设计的系统包含四个层面，每一层解决一个特定的问题：这些层面相互依赖。知识系统的文件结构使自动化编辑成为可能；推理层的明确程序使故障归因变得可追踪；评估框架对每一次变更进行把关；改进循环则反馈到知识和推理两个层面。移除其中任何一层，其他层面都会退化。

Building the Organizational Second Brain

构建组织的“第二大脑”

Large organizations can accumulate thousands of documents as a byproduct of expert work. It is tempting to treat those documents as organizational knowledge, but the real knowledge is implicit: how experts reason, what they prioritize, and how they resolve ambiguity. An agent that retrieves document chunks at inference time has to re-derive that reasoning from raw sources on every run, which is slow, error-prone, and inconsistent. We make that implicit knowledge explicit ahead of time. 大型组织在专家工作中会积累数以千计的文档。人们很容易将这些文档视为组织知识，但真正的知识是隐性的：专家如何推理、他们优先考虑什么，以及他们如何解决模糊性。一个在推理时检索文档片段的智能体，必须在每次运行时从原始来源重新推导推理过程，这既缓慢又容易出错且不一致。我们提前将这些隐性知识显性化。

A long-running offline process reasons through source documents and distills them into structured knowledge files – curated statements of how the organization interprets its domain, with constraints, boundaries, and routing implications made machine-readable. Most significantly, that knowledge then forms the basis of a feedback loop that allows the agent to learn from and implement feedback from human experts without the underlying model having to be retrained. 一个长期运行的离线进程会对源文档进行推理，并将其提炼为结构化的知识文件——即关于组织如何解读其领域的精选陈述，其中包含机器可读的约束、边界和路由含义。最重要的是，这些知识随后构成了一个反馈循环的基础，使智能体能够学习并实施人类专家的反馈，而无需重新训练底层模型。

The industry has converged on a similar idea. Andrej Karpathy’s LLM Wiki structures agent knowledge as a navigable graph of files, and Google’s Open Knowledge Format standardizes this for cross-agent interoperability. The shared insight is that knowledge should be pre-extracted, explicitly structured, and progressively disclosed rather than re-derived on every query. 业界已经汇聚到类似的理念上。Andrej Karpathy 的 LLM Wiki 将智能体知识结构化为可导航的文件图谱，而 Google 的 Open Knowledge Format 则将其标准化以实现跨智能体的互操作性。共同的见解是：知识应该被预先提取、显式结构化并逐步披露，而不是在每次查询时重新推导。

We extended these principles into a system where citation fidelity and institutional consistency are non-negotiable, organizing 200+ files into a strict taxonomy: 我们将这些原则扩展到一个引用保真度和组织一致性不可妥协的系统中，将 200 多个文件组织成严格的分类法：

• Position files capture authoritative organizational stances: how the organization has decided to interpret a given domain question, along with its constraints, boundary conditions, and machine-actionable routing implications that tell the reasoning layer when to apply it. 立场文件：捕捉权威的组织立场，即组织决定如何解读特定的领域问题，以及其约束条件、边界条件和机器可执行的路由含义，这些含义会告诉推理层何时应用该立场。
• Taxonomy and vocabulary files act as an authoritative glossary for the terms the organization uses to describe its domain, such as entity types, activity categories, and classification tiers. Each is maintained as a single source of truth so the agent and the organization use language consistently. 分类法和词汇表文件：作为组织描述其领域所用术语的权威词汇表，例如实体类型、活动类别和分类层级。每一项都作为单一事实来源进行维护，以确保智能体和组织在语言使用上保持一致。
• Routing indexes map input characteristics to the relevant positions and procedures, determining which files apply without relying on embedding similarity alone. This makes retrieval deterministic and auditable. 路由索引：将输入特征映射到相关的立场和程序，在不单纯依赖嵌入相似度的情况下确定适用哪些文件。这使得检索过程具有确定性和可审计性。
• Gateway files define threshold tests the agent must pass before entering an analytical domain, preventing it from applying specialized knowledge where it does not belong. Every file declares its dependencies (depends_on) and consumers (referenced_by) in YAML frontmatter, forming a bidirectional… 网关文件：定义了智能体在进入分析领域前必须通过的阈值测试，防止其在不适用的地方应用专业知识。每个文件都在 YAML 元数据中声明了其依赖项（depends_on）和使用者（referenced_by），从而形成了一个双向的……