2026
- Weekly Review — Nov 9–15, 2026
TL;DR: Five posts this week. Monday tests TabPFN, a foundation model for tables, against XGBoost on four public datasets. Wednesday reads the docs of the new Well-Architected Agent, which ranks …
- Many Agent Steps Are Classification in Disguise
TL;DR: Routing a request, deciding whether a human should confirm, and rating urgency are decisions over options you already know. Agents still often hand them to a large model, which writes a …
- A Foundation Model for Spreadsheets
TL;DR: For about twenty years, gradient-boosted trees, in recent years mostly XGBoost and CatBoost, have been the default for tabular prediction, and deep learning kept losing to them. TabPFN takes a …
- Compression Is Intelligence
TL;DR: Information theory says prediction and compression are the same thing measured two ways. That gives the cleanest mental model I know for what a large language model does: it is a compressor of …
- Intelligence Is Collective, Not Artificial
TL;DR: Michael I. Jordan, the man Science magazine named “the world’s most influential computer scientist,” has never considered himself an AI researcher. His thesis: intelligence is not a property of …
- Your AI Judge Needs a Judge
TL;DR: Most teams I see ship LLM judges without testing them against human labels. The result: judges that are confidently, consistently wrong on 30%+ of cases. Hamel Husain’s “critique shadowing” …