Skip to content

Converting messy documents into structured data with Peter Staar of Docling

Play

In this episode, I speak with Peter Staar of Docling about the wonderful open-source project that converts messy documents into structured data, perfect for ingestion into AI tools and more. Recorded live at PyTorch con in 2026.

Published: 27 June 2026

Join my mailing list

My roughly fortnightly newsletter, sent via Substack.

Latest issue: Don't let the Sun go down on Open Source (24/09/2026)

View archive