For M&A analysts, project finance, and transaction advisory

Understand the model. Before you present it.

Navigate any complex financial model in hours — M&A diligence, project finance stress-tests, infrastructure analysis. Every formula traceable. Every scenario exportable to a clean .xlsx your committee can review.

No credit card · No data retention by default · Export to standard .xlsx

The problem

You received the model. The committee meets next week.

Whether it's a sponsor's LBO or a borrower's project finance model, the situation is the same: a file with too many tabs, assumptions buried in cells nobody documented, and a deadline that does not move. You paste fragments into Claude to gain ground — but the agent can't see the whole structure, you re-explain every session, and the data sits in a personal AI account.

M&A due diligence

The model arrived Friday. The IC memo is due Thursday.

15 tabs, 3,000 rows, variable names that made sense to whoever built it in 2019. You need to spot the key assumptions, identify where the leverage actually sits, and produce a view your MD can take to the investment committee — fast. Layerz reads the structure your agent can't see, so the analysis works on logic, not cells.

Project finance & infrastructure

A model arrives from the borrower. The credit committee votes next week.

20 tabs, 6 debt tranches, DSCR computed somewhere in a column you have not found yet. Layerz maps debt sizing, DSCR logic, cover ratios — your agent stress-tests revenue ramps, construction cost overruns, rate shocks against the actual financial structure, not against a scan of the grid.

What Layerz changes

From cells to financial logic.

TodayWith Layerz
Re-explain the model every AI sessionModel structure persists — your agent knows the architecture
AI reads cells, not financial logicVariables, formulas, and dependencies are machine-readable
Break-even, DSCR, cover ratios are manual recalcsRun sensitivity on any assumption — leverage, revenue ramp, rate shock
Model audit is a manual trace through linked cellsEvery formula traced to its source — defensible in front of an auditor
Building under deadline means recycling a 2019 fileDefine the structure once — reinstantiate for each new deal
Upload client or project data to a personal AI accountBYOA: your tokens, your environment — data does not leave your setup
The committee and the counterparty need .xlsxExport a clean standard file — no Layerz login required on their end
How it works
1

Hand the model to Claude

Drop the received Excel in your Claude session and tell it to map the model. Claude calls the Layerz MCP to extract the structure — debt tranches, DSCR logic, cover ratios, equity waterfall — into a typed backend it can navigate.

2

Stress-test in conversation

Ask Claude in plain English to find undocumented assumptions, run 5×5 sensitivity tables, or flag where cover ratios break. It works against the structure, not against a scan of the grid.

3

Export when the memo is ready

Tell Claude to export. Your IC, your MD, your co-lender opens a clean .xlsx. The structure persists across sessions, so you reopen tomorrow and continue.

Talk to your agent

One line. Claude does the rest.

  • map this LBO and find where leverage breaks
  • run a 5×5 sensitivity: DSCR × revenue ramp
  • find every undocumented assumption in the debt schedule
  • export the base case as a clean xlsx for the committee

Run these in any Claude session with the Layerz MCP installed.

For the deal you did not originate

Most diligence is reading someone else's model.

Most transaction 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.

For the deal you originate

Templates that hold up across every transaction.

On the build side, the wall is not the model itself — it is rebuilding it from a 2019 file every time the next deal lands. PPP, renewable, toll road, real estate development, LBO: same logic, slightly different shape.

Layerz lets you define the structure once and reinstantiate for each new deal. Every variable named, every formula traceable, every change logged. The model passes audit and the committee gets a clean export.

Define once. Reinstantiate per deal. Defensible end-to-end.

What you can do

In the time you have, not the time you’d like.

Map a received model in under two hours

LBO, PPP, infrastructure, M&A: financial architecture, debt tranches, DSCR logic — all mapped before the kickoff call.

5×5 sensitivity tables

Leverage vs. revenue ramp. Construction cost overrun vs. rate shock. Clean output ready for the committee.

Spot what the sponsor's advisor hid

Circular references, undocumented assumptions, formula inconsistencies — flagged before the committee meets.

Reusable deal templates

PPP, renewable energy, toll road, real estate, LBO. Define once, reinstantiate per deal.

A model that passes audit

Every variable named, every formula traceable, every change logged. Defensible end-to-end.

Map the model. Stress-test. Ship the memo.

No credit card · No data retention by default · .xlsx export always free