# Harvey — Enterprise-Grade RAG Systems

- Company: Harvey (harvey.ai)
- Announced: 2025-02-20T23:16:24.389000+00:00
- Category: not stated
- Coverage: not counted
- Announcement: no
- Group: routine
- Source: https://www.harvey.ai/blog/enterprise-grade-rag-systems
- Record: https://forck.live/items/17715-enterprise-grade-rag-systems
- Subject: Harvey

Harvey published a technical guide explaining how it builds enterprise-grade Retrieval-Augmented Generation (RAG) systems for legal and professional services firms. The post covers the three types of data sources in its platform, the challenges of enterprise retrieval, and the criteria for selecting vector databases, including a comparison of Postgres with PGVector and LanceDB. Harvey requires a vector database that can be hosted within a private cloud environment and store embeddings and source data in customers' own secure storage.

## Evidence

Verbatim from https://www.harvey.ai/blog/enterprise-grade-rag-systems:

> At the heart of RAG systems are vector databases (VectorDBs), which enable efficient searching across massive datasets. Choosing the right database requires careful consideration of performance and compliance.

---

Record: https://forck.live/items/17715-enterprise-grade-rag-systems
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md