01Multi-Agent News Pipeline
Fund Analyst News Pipeline
A multi-agent news pipeline for a ten-stock watchlist. It polls financial news and exchange filings around the clock, filters the noise cheaply, and for the handful of stories that survive writes a bull case and a bear case against the fund's own thesis. It stops there on purpose: a considered view, not a recommendation.
- Problem
- A markets feed for even a ten-stock watchlist is mostly noise, and running every article through an LLM doesn't scale or build the kind of trust a fund manager needs to weigh a call.
- Approach
- String, fuzzy, and embedding entity matching filter the feed before any model call runs. A DIKW-staged pipeline scores each surviving story's materiality against the company's own thesis card, then runs an adversarial two-sided debate, with every call logged and cached.
- Proof
- 439 articles fetched · 112 duplicates collapsed · 11 events surfaced, across 12 live sources
- voyage-finance-2 separated duplicate reports from unrelated pairs by a +0.214 margin, the widest of three embedding models tested
- Every LLM call cached by input hash; a full rerun costs $0 and needs no network
