The second catch is that the entire approach rests on the assumption that language models are tireless and accurate librarians, but in practice, language models are bad at writing clean markdown at scale—they botch the formatting, mangle headers, and invent links to files that were never created.

causalpending

Speaker

Unidentified Speaker — Google's OKF: Why a Folder Beats the Vector Database [P_E29-87THI]

Evidence Quote

“In practice, language models are bad at writing clean markdown at scale. They botch the formatting, mangle headers, invent links to files that were never created.”

Source

Google's OKF: Why a Folder Beats the Vector Database— Devsplainers
Created: 8/13/2026, 7:46:31 AM

Your notes

Loading notes...