data-baker

What problem does it solve?

Data Baker automates the generation of realistic synthetic datasets for AI agent training, testing, and Retrieval-Augmented Generation (RAG) workflows by turning meeting transcripts, requirements documents, and process flows into rich, cross-referenced data with proper metadata and formatting.

Core Features & Use Cases

• 9 data generation strategies aligned with common data needs (organizational knowledge, identity data, professional documents, financial records, medical data, legal contracts, product catalogs, customer data, and email/communication)
• Batch generation with master entity registries to ensure consistency across documents and support networked data
• RAG-optimized frontmatter, sectioning, and cross-document references for seamless retrieval
• Flexible input sources (meeting transcripts, requirements docs, process flows, or plain text descriptions) to generate domain-appropriate synthetic data

Quick Start

Specify domain, key entities, and document types, then run the generator to produce a starter synthetic dataset.