---
title: "An Organizational Second Brain: Building an AI That Learns From Experts"
description: "Organizations often retain specialist knowledge in experts’ heads, leaving routine compliance assessments slow, inconsistent, and difficult to scale. The proposed “second brain” combines a structured, auditable knowledge architecture with a separate reasoning layer and a self-improvement loop that converts expert feedback into verified, regression-tested text updates without model retraining. High-density organizational positions, taxonomies, routing indexes, and gateway files are maintained in a dependency-linked filesystem, while sparse supporting material remains available through semantic or lexical retrieval. Composable recipes encode expert procedures, and an evaluation pipeline proposes auditable pull requests, retries failed fixes, and expands regression coverage after approval. After six weeks, SMEs reportedly found outputs useful almost always, assessments fell from days to minutes, and improvement cycles produced zero regressions; the architecture is presented as applicable beyond compliance."
---

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

[Meta](https://yomu.fyi/company/meta) · By Shaurya Sengar, Jason Nawrocki, Jay Shah, Prashant Kommireddi · Sep 2, 2026

**Type:** Problem & solution

## Summary

Organizations often retain specialist knowledge in experts’ heads, leaving routine compliance assessments slow, inconsistent, and difficult to scale. The proposed “second brain” combines a structured, auditable knowledge architecture with a separate reasoning layer and a self-improvement loop that converts expert feedback into verified, regression-tested text updates without model retraining. High-density organizational positions, taxonomies, routing indexes, and gateway files are maintained in a dependency-linked filesystem, while sparse supporting material remains available through semantic or lexical retrieval. Composable recipes encode expert procedures, and an evaluation pipeline proposes auditable pull requests, retries failed fixes, and expands regression coverage after approval. After six weeks, SMEs reportedly found outputs useful almost always, assessments fell from days to minutes, and improvement cycles produced zero regressions; the architecture is presented as applicable beyond compliance.

## Context

Specialist knowledge is often implicit in experts’ heads, while large organizations face repeated questions, lengthy manual research, inconsistent assessments, and limited expert capacity. General-purpose LLMs also lack the institutional context needed to distinguish general possibilities from an organization’s established positions and priorities.

## Approach / What changed

The system separates structured domain knowledge from reasoning procedures, using a dependency-linked filesystem of positions, taxonomies, routing indexes, and gateway files alongside composable recipes. Sparse supporting material is accessed through semantic or lexical retrieval. Expert feedback enters an automated compilation and evaluation pipeline that produces audited text diffs, retries failed changes, and adds approved fixes to regression tests without retraining the model.

## Takeaways

- The knowledge system uses YAML frontmatter with depends\_on and referenced\_by fields to form a bidirectional dependency graph, enabling impact tracing when files change.
- High-density, frequently used organizational knowledge is stored in curated wiki files, while sparse or situational material remains in semantic or lexical retrieval systems.
- After three development sprints over six weeks, the system reported assessments reduced from days to minutes, useful outputs almost always, and zero regressions across improvement cycles.

**Tags:** [AI Agents](https://yomu.fyi/topic/ai-agents), [Architecture](https://yomu.fyi/topic/architecture), [Human-in-the-Loop](https://yomu.fyi/topic/human-in-the-loop), [Knowledge Graph](https://yomu.fyi/topic/knowledge-graph), [Retrieval-Augmented Generation](https://yomu.fyi/topic/retrieval-augmented-generation)

- Source: [Meta](https://engineering.fb.com/2026/09/02/ml-applications/organizational-second-brain-ai-learns-from-experts)
- Source URL: https://engineering.fb.com/2026/09/02/ml-applications/organizational-second-brain-ai-learns-from-experts
- Ingested by Yomu: 2026-09-03T00:00:51.673Z

[Read original post](https://engineering.fb.com/2026/09/02/ml-applications/organizational-second-brain-ai-learns-from-experts)