Building a Transparent AI Pipeline: 59 Weeks of Automated Political Scoring with Claude API

A developer built an automated AI pipeline running for 59 weeks that ingests news articles, clusters them using Claude Haiku, and scores political events on two axes—constitutional damage and media distraction—using Claude Sonnet. The system detects 'smokescreen' patterns where high-distraction events coincide with under-covered institutional damage, with full methodology published for transparency. Cost optimization through dual-model architecture keeps the pipeline at ~$30/month.