video1.0<iframe src="https://www.loom.com/embed/f538f79d328f420aac248f0383ebb3ee" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/f538f79d328f420aac248f0383ebb3ee-deafdfc6d5bad2d3.gif244.815This Loom presents an earnings sentiment project that correlates CEO confidence in earnings call transcripts with post-earnings stock movement to support short-term trading decisions. It shows company-level analytics for one-day, three-day, and seven-day returns after earnings calls, including predictability when CEO confidence correlations are stronger for more CEO-driven names like Elon Musk. The dashboard provides historical average confidence by post-seven-day return, a feed and company inspection with deeper analysis via a Claude API rundown of three relevant transcript phrases, and a leaderboard and sector view. The architecture pulls data from sources like WiFi Ads for prices plus Alpha Vantage, SEC EDGAR, and FMP, then processes it through Python ingestion to Kafka topics, a correlation scoring service, MongoDB, and a BFF in ExpressJS with a Nextcloud front end, including email notifications for updates.