Article & Topic Classifier

An aggregation engine ingesting at volume is worthless if everything lands in one undifferentiated feed. Relevance has to be decided at ingest, not by the reader.

A classification layer that categorises and tags incoming articles by topic and relevance and generates summaries automatically, routing content to the right destination and feeding the automated email sends. Built on Claude Haiku deliberately — classification is a high-volume, low-reasoning task where the cheaper, faster model is the correct engineering choice, and paying Sonnet rates for it would be a design error.

Lets the platform run continuously without a human reviewing every item, while keeping the feed coherent enough to be worth subscribing to.