Product
AI Hedge Fund Research Assistant
Research notes and investment theses that record what you believe, why, and what would prove you wrong.
What it does
Portfolios and positions
Holdings with cost basis and quantity, denominated explicitly, so exposure is computed rather than estimated.
Research notes
Notes tied to a security, with sources recorded, so a claim written six months ago can still be traced to what it was based on.
Investment theses
Stance, conviction, key assumptions, and risks as structured fields. The assumptions are what you revisit when the position moves against you.
Fundamental analysis
Ratios and trends computed deterministically from imported financials, with the inputs shown alongside the output.
Comparables
Peer comparison on a consistent basis, with the normalisation choices stated rather than buried.
Cited answers
Questions about a holding are answered from your own research and imported data, with citations — not from the model recalling a company.
How answers are produced
Every figure comes from a deterministic calculation against your imported data, returned with its inputs and a step-by-step trace. The assistant chooses which calculation to run and explains the result; it never computes the number itself.
Where a question cannot be answered from the data you have loaded, the answer says what is missing rather than estimating.