Notes
Blog
Practical notes on evidence synthesis, inspectable AI workflows, and the checkpoints that keep reviews defensible.
4 Sept 2026
LLMs invent MeSH and Emtree — look headings up
Language models are useful for synonym harvest and field-tag drafts. Controlled vocabulary headings are not something a model should invent — look them up, then approve.
Read note
28 Aug 2026
Starting AI‑assisted systematic reviews
How AI can support systematic reviews while keeping decisions auditable, challengeable, and reproducible.
Read note