TableAI article
Designing an evidence-first AI catalog
How canonical records, multi-type classification, and source-backed benchmark claims shape this catalog.
Articles
Methods, systems, and source material.
Original notes live beside reviewed external publications, each with a clear canonical source.
TableAI article
Operating agent infrastructure on Cloudflare Workers
A practical account of preserving health checks, scheduled collection, D1 history, and protected operations APIs.
External publication
Introducing the Model Context Protocol
An introduction to an open protocol for connecting AI assistants with tools, systems, and data sources.
External publication
SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
The research paper introducing a benchmark based on repository issues, code snapshots, and executable tests.