Use case
Financial models, built right.
Tell your AI coworker the company and the model. You get a real Excel file with =SUM, =NPV, and =IRR formulas — not numbers the model typed. Sensitivity tables, scenario tabs, and footnotes included.
What it does
From "build me a DCF" to a working Excel file.
Your AI coworker pulls the latest filings, derives historical financials, and projects forward with assumptions you can audit. The output is a working Excel file — formulas live, links between tabs intact, sensitivity tables already wired up.
Models supported out of the box: discounted cash flow (DCF), leveraged buyout (LBO), three-statement, comparable companies (comps), precedent transactions, returns analysis, and unit economics.
How it's different
Models the way analysts build them.
Real formulas, not numbers
Cells contain =SUM, =NPV, =IRR. Change a driver and the model recalculates. ChatGPT types numbers; we wire formulas.
Auditable assumptions
Every projection links back to a source — a filing line, an analyst estimate, a guidance call. You can see why a number is what it is.
Sensitivity built in
Two-variable sensitivity tables on WACC × terminal growth, EBITDA × multiple, etc. — generated alongside the base case, not as an afterthought.
What you say to it
One-line briefs, real outputs.
"Build a DCF for ABC Corp with 2031 horizon."
Three-statement linked, WACC computed, terminal growth + multiple toggle, two-variable sensitivity tab.
"LBO the take-private at $42/share, 6.5× leverage."
Sources & uses, debt schedule, returns waterfall, IRR / MOIC sensitivity to exit multiple and hold period.
"Comps for the consumer-tech peer set."
EV/Revenue, EV/EBITDA, P/E across the peer set with means, medians, and quartiles auto-computed.
Pull up a desk.
500 free credits. No credit card. Your AI coworker is ready whenever you are.