LLMFSD: LLM Fake Structured Data, faking structured data from any LLM

Software

Proposed an innovative approach to generate fake structured data using LLMs. The project enables developers to create synthetic datasets via SQL-like queries, bridging the gap between unstructured LLM outputs and structured data requirements for testing and prototyping. Developed a Python package (llmfsd) available on PyPI.