The Machine in the Library

Notes from building and running an on-premise AI inference machine - hardware quirks, pipeline design, thinking and rethinking in the open.

Posts

• Introducing Paratext 9th July 2026 - Making an images-to-output workflow that is simple, extensible, portable, and intuitive.
• Reaching good enough 25th June 2026 - Reaching a 'good enough' threshold on model-generated metadata
• Fine tuning a model on the local machine - powered by wind 5th June 2026 - Fine-tuning a model entirely on the local machine, and making a tool to do this as sustainably as possible.
• Why the Framework Desktop 4th March 2026 - Why we chose the Framework Desktop as our local AI platform.
• Beginning the work 18th February 2026 - How we think about AI, and the beginning of a local journey.