Navigate any project finance model in hours — debt sizing, DSCR, equity waterfall, sensitivity tables. Every formula traceable. Every scenario exportable to a clean .xlsx your credit committee can review.
No credit card · No data retention by default · Export to standard .xlsx
20 tabs, 6 debt tranches, DSCR computed somewhere in a column you haven’t found yet. You need to understand the key assumptions, flag the stress scenarios where cover ratios break, and produce a credit memo — by end of week.
You paste what you can into Claude. It helps — but it can’t see the whole structure at once, and you re-explain the same thing every session. The logic doesn’t persist. Every morning starts from scratch. And you’re doing this on your personal AI account, with project data.
Layerz is the structural layer between your model and your AI. It reads the variables, formulas, and financial logic — not just the cells. Your agent navigates the model, stress-tests assumptions, and flags where cover ratios break — while the data stays in your environment.
| Today | With Layerz |
|---|---|
| Re-explain the model every AI session | Model structure persists — your agent knows the architecture |
| AI reads cells, not financial logic | Variables, formulas, and debt relationships are machine-readable |
| DSCR and cover ratios are manual recalculations | Run sensitivity on any assumption — debt sizing, revenue ramp, cost overrun |
| Building under deadline means recycling a 2019 file | Define the structure once — reinstantiate for each new deal |
| Model audit is a manual trace through linked cells | Every formula traced to its source — defensible in front of a lender’s auditor |
| Upload project data to a personal AI account | BYOA: your tokens, your environment — data does not leave your setup |
| The client and the credit committee need .xlsx | Export a clean standard file — no Layerz login required on their end |
Upload the Excel. Layerz maps the structure: revenue assumptions, cost lines, debt tranches, DSCR logic.
Your Claude or Codex reads the model structure directly. No copy-paste, no re-explanation session to session.
Ask your agent to identify where cover ratios break under revenue stress, run sensitivity on construction costs and interest rates, flag undocumented assumptions.
On the origination side, define the structure once. Reinstantiate it for the next greenfield or brownfield.
Standard .xlsx. Your credit committee, your MD, your co-lender opens a spreadsheet. Layerz is not visible.
Most project finance work is not building models from scratch. It is understanding a model someone else built, under assumptions that were in their interest to present favorably.
Layerz gives your AI the structural context to read that model intelligently — not just surface-level. When the sponsor’s advisor says the base case DSCR is 1.35x, your agent already knows where the revenue growth assumption is, what the debt service schedule looks like, and what stress scenario makes it break.
That is a different quality of diligence than pasting tab by tab into a chat window.
LBO, PPP, infrastructure: financial architecture, debt tranches, DSCR logic — all mapped before the kickoff call.
Construction cost overrun vs. revenue ramp. Clean output ready to present to the credit committee.
Circular references, undocumented assumptions, formula inconsistencies — before the committee does.
PPP, renewable energy, toll road, real estate development. Define once, reinstantiate per deal.
Every variable named, every formula traceable, every change logged. Defensible end-to-end.
€0 /month — forever
Unlimited deal models — your data stays in your environment, free as long as you're the only editor
€29 /editor/month excl. tax
For when the deal team edits the same models
No credit card · No data retention by default · .xlsx export always free
Try it free — bring your own Claude