Fraud detection is an engineering problem.

bearsignal.ai — Launch Manifesto (also serves as site homepage)

For fifty years, forensic accounting research has produced models that work: Benford’s Law deviations, discretionary accruals, accrual quality scores, structural distance-to-default. The literature is public. The mathematics is sound. And almost nobody runs it — not systematically, not across entire markets, not every reporting period.

Why? Because running forensic mathematics at market scale was never a research problem. It’s a data engineering problem, a calibration problem, an infrastructure problem. Academia doesn’t build pipelines. Funds build them for their own book and tell no one. Sell-side has no incentive to find fraud at all.

BearSignal exists to close that gap. We are building the infrastructure to run published forensic research — and our own extensions of it — against every company listed in the U.S. and Hong Kong, continuously.

This site is where we write about how. The detection mathematics and what the literature gets right. The pipelines that turn SEC XBRL and HKEX disclosures into clean, comparable fundamentals across two regulatory regimes. The architectural decisions — some obvious in hindsight, some learned expensively — behind a system where statistical evidence and human judgment are kept deliberately, structurally apart.

We write under our real constraints: a small team, a production system, real capital decisions downstream of our output. We name our tools when naming them is useful. We show our reasoning when the reasoning is the point. What we don’t publish is the judgment layer itself — that’s the product, and you wouldn’t respect us if we gave it away.

If forensic mathematics at market scale sounds like a problem worth your career, we’re hiring — in Silicon Valley, in person, for people who want to build things that find what’s hidden.

→ Engineering · Methodology · Careers