An Organizational Second Brain: Building an AI That Learns From Experts
Meta built an agent that turns domain experts into institutional memory — and makes their knowledge auditable.
Why it matters
An enterprise agent architecture that preserves specialist knowledge at scale while maintaining auditability. This is a deployment pattern (knowledge layer + agent reasoning) that practitioners building internal knowledge systems will study.
The key facts
10 to knowMeta engineering blog post on organizational AI agent
Two-layer architecture: structured knowledge + auditable reasoning
Designed for domain expertise preservation and organizational access
Agent learns from experts, not from raw data alone
Published September 2026
Meta Engineering built domain-specific agent with structured, auditable knowledge architecture
Two-layer integration: knowledge layer + expert learning layer
Framed as organizational knowledge preservation and access tool
Published September 2, 2026 on Meta Engineering blog
Focus on enterprise knowledge management use case
Go to the source
Meta Engineeringengineering.fb.com
Publisher excerpt: 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…