Sheaf
Portfolio IQ
A worked book of 180 exposures across five segments. Sample data, the same engine. Open the full demo.
What it does
The agents do the work a credit team does, case by case and across the whole book. Before any of them acts, Sheaf checks that the reasoning holds together.
Income, commitments, bureau score, bank statement. Contradictions between sources are found and named, not averaged away.
A quote that rests on a disputed fact stops before it goes out. The record shows the rate, the facts behind it, and what was checked.
Concentration, stress, drift, the cases whose facts have moved. Agents propose actions on the book, a floor, a provision, early help, and each one goes through the same check and the same gate.
How it works
Underneath Portfolio IQ is the Sheaf engine: a mathematical measure of whether the facts an agent is about to act on agree with themselves. If they contradict, the agent stops.
Before an agent acts, Sheaf measures whether its reasoning holds together: the facts it is acting on, the rules it is under, and what the other agents concluded. The measurement comes from published mathematics, not from asking a model whether it is sure.
An action that fails the check is not lost and not forced. It is parked, with the disagreement laid out in plain words, until a named person on your team releases it or turns it down.
Every check, every hold and every release is signed and timestamped by an independent authority. When the regulator asks why, you hand over the record.
Proof
See it run on a book. Take the record away with you. And know that the check itself has been examined by people who do not work for us.
Move the shock, watch the bands shift, pick the customer in the red and read their file. Then watch the agents propose actions on the book and see which ones Sheaf holds. Sample data, the same engine that runs on yours.
For every decision: the facts it rested on, the check that ran, whether it was held, who released it and why. Signed, and timestamped by an independent authority, so it stands on its own after the fact.
The method behind the check is a published paper with a permanent DOI and a provisional patent. Its central theorems are formalised in Lean and checked by machine, and the files are public. Anyone can run them.
Make and break Sheaf
Bring us your most challenging business problem that you want agentic AI to solve.Anonymise it. We'll build it, stress it and have it ready for deployment. Ten businesses only.