Filing-grounded reasoning
Reads the 10-K/10-Q/8-K trail for every claim. Quotes are pulled live from Finnhub; no LLM-generated prices ever reach the dossier.
All use cases
Equity researchLive in portal
AI Advisor 2.0
Build the thesis on any US stock. Reads filings, valuations, and macro context — and tells you when the data revises the call.
AI Advisor 2.0 is a deterministic-first equity advisor that grounds every claim in a verifiable source. It reads SEC filings (10-K, 10-Q, 8-K), pulls live quotes, runs a sector-equivalent reverse-DCF valuation, layers in macro context, and produces a structured thesis with explicit revision triggers — so when the next quarter's data flips a key assumption, you know exactly what changed and why.
What makes it different: Deterministic-first reasoning. Every number traces to its source filing or quote API. The model never invents prices.
What it does
Capabilities
Sector-equivalent valuation
Routes financials to ROE inversion, REITs to AFFO growth, banks to dividend growth — not a one-size-fits-all DCF that pretends every company is the same.
Macro context integration
Cross-references your thesis against the Macro Triggers regime + sector outlook so you see when a stock-specific call collides with a macro headwind.
Revision triggers, not point-in-time
Output names the specific data points that would flip the call. When the next quarter prints, you see whether the thesis is intact or needs review.
How it works
Input → output
Input
DHR · Danaher Corporation
- 1Pull last 4 quarters of 10-Q + most recent 10-K from EDGAR (XBRL-grounded)
- 2Compute valuation inputs: cash, debt, shares outstanding, FCF, ROE — with provenance on each
- 3Run reverse-DCF: solve for implied FCF growth at current price
- 4LLM synthesizes the qualitative thesis using ONLY the deterministic inputs above
- 5Surface revision triggers: 'thesis breaks if X falls below Y' or 'confirms if Z prints above W'
Output
A structured thesis card: verdict (Hold / Add / Trim), confidence level, 3-5 cited key signals, and 2-3 named revision triggers tied to specific data points.
Output preview
portal · live
DHR
Danaher Corporation · life sciences
Hold · thesis intact
Key signals
- FCF yield3.8%10-K §Item 7
- Implied growth4.2% / 7yReverse-DCF
- Insider activitynet +$2.1M (180d)Form 4
Revision triggers
- ⚠Trim if Q4 FCF falls below $1.4B → implies growth re-rating
- ↑Add if life sciences segment margin expands above 22% → confirms operating leverage thesis
Source: EDGAR · Finnhub · provenance traced
Why this isn't a wrapper
Why this isn't another GPT wrapper
Most AI equity tools let the model freely write the analysis — including the numbers. Ours doesn't.
- Numbers are computed deterministically by code. The LLM provides the narrative; the math is verifiable.
- Every metric on a dossier carries provenance — click any number to see the source filing/cite.
- When the data doesn't support a claim, the model abstains. You see '—', not a plausible-looking guess.
- Mechanism-only signals (cited but not statistically validated) are labeled as such — never dressed up as conviction.
Want to try it?
Book a 30-minute walk-through. We'll demo a real run on a ticker you pick.