On time, memory, and what large language models cannot do with either It started, as these things often do, with a podcast. I was listening to Orit Halpern on the BBC’s ‘In Our Time‘. She is an historian and self-declared cyberneticist, sharp on the politics of data. One of her comments caught my attention. She was talking about how computational …
Adding Quality Control to Andrej Karpathy’s LLM Wiki
On April 2nd, Andrej Karpathy posted on GitHub and X about building a Persistent LLM Wiki – a structured knowledge base that compounds over time, where every paper you read, every idea you encounter, gets added into an interconnected graph of notes. The tool he reached for was Obsidian. The idea resonated immediately. Obsidian’s graph view, its bidirectional links, and …


