Burrow

A self-hosted data ingestion pipeline and retrieval API that streamlines and accelerates the development of RAG applications

Automated Ingestion Pipeline

Ease of getting documents into a database via both friendly GUI and RESTful endpoints, the pipeline takes care of the rest.

Intelligent Document Processing & Retrieval

Layout-aware document processing and structure-preserving chunking, combined with an API offering hybrid search, reranking, and metadata filtering.

Parsing
Parsing
Chunking
Chunking
Embedding
Embedding
Vector Store
Vector Store
Hybrid Search Reranking Metadata Filtering
Burrow CLI Screenshot

Simplified AWS Deployment Using the Burrow CLI

Automated deployment to your AWS environment using a self-hosted, serverless architecture where performance and cost scale based on usage patterns.

Next Steps

Ready to dive deeper into Burrow? Explore our comprehensive case study or download from GitHub.