Institutional CRE underwriting
Underwrite
Toggle navigation
Cayuga Commons | From chaos to conviction

From chaos to conviction.

A broker sentence becomes a defensible IC memo. Scroll: the deal is detected, modeled, stressed, and decided, and every number can be traced back to the sentence that began it.

Underwrite a dealOpen the worked example
  1. Source
  2. Model
  3. Risk
  4. Decision
01 | Source

A broker sentence becomes evidence.

“24 residential units at $2,400 per month. 4,200 SF of retail at $38 per SF. $8.5 million asking price.”

The parser reads the memo and quotes its evidence: every extracted value carries the verbatim source sentence, so nothing enters the model without a citation.

02 | Model

Assumptions snap into a model that recomputes.

Purchase price
$8,500,000
Senior debt
65% LTV
Rent growth
3.00%
Exit cap
6.50%

The detected values snap into a deterministic model. Pure TypeScript carries the pro forma, debt schedule, and returns; the same inputs always produce the same outputs, with no run button and no hidden save step.

03 | Risk

One thousand trials, and the stress that matters.

P(IRR at or above 15%)
23.70%
Downside IRR at 7.00% exit
13.51%
Minimum DSCR
1.46x
Covenant floor
1.25x

One thousand seeded trials, DSCR stress against the covenant, and 143 deterministic modules with zero LLM involvement in the math path. The model becomes a risk environment before it is allowed to become a recommendation.

Inspect the engine source
04 | Decision

PURSUE. Every number returns to its sentence.

Base caseExit cap 6.50%
Project IRR
15.34%
Verdict
PURSUE

The deal clears the 15% return threshold and debt service remains covered. Raise the exit cap to 7.00% in the tool and the engine argues back.

From broker email to defensible IC memo. Sources remain traceable at every step.

The first pass through a CRE deal can be fast without becoming disposable.

Underwrite turns source material into a reviewable decision. AI helps read the broker file. Deterministic code carries every return, coverage ratio, stress, and verdict.

The product is built for sponsors, acquisitions analysts, and LPs who need to move quickly, then show exactly how they arrived. The memo is not a summary pasted on top of the work. It is the work, organized for review.

The model should
survive the meeting.
Process

A broker sentence becomes an IC decision.

  1. 01

    Paste the broker IM

    Drop in a memo, marketing flyer, or rent roll. The AI parser extracts purchase price, NOI, debt assumptions, and tenant detail with source quotes you can verify.

  2. 02

    Interrogate the assumptions

    Override any extracted value. The deterministic engine recomputes returns, debt coverage, sensitivities, and the verdict as you type.

  3. 03

    Carry the decision into IC

    Open the memo, trace each metric, export the workbook, or keep the workflow local with Confidential Mode. The audit trail stays attached.

Proof, not posture

The output arrives with receipts.

Open the source, inspect the formula path, and read the memo in the order an investment committee would challenge it.

392
Regression assertions
Canonical outputs locked
143
Live Phase 3 modules
Computed deterministically
0
LLM calculations
In the return path

Why the trail holds

Reviewable engine source
The methodology links to the TypeScript that produces the math. The same inputs produce the same outputs.
Click-to-verify extraction
Every AI-populated field retains the verbatim source sentence that supported it.
Confidential Mode
Turn off parser and sharing flows while the deterministic engine continues in the browser.

Inside the IC memo

143 modules live
  1. 01Deal Strength Score
  2. 02Monte Carlo Distribution
  3. 03Sponsor Co-Invest Sweep
  4. 04Stress Test Matrix
  5. 05Lender Risk Rating
  6. 06Refinance Rate Risk
  7. 07Capital Stack Visualization
  8. 08GP Promote Yield
  9. 09Reserve Adequacy
  10. 10Sale vs Refinance Decision
  11. 11Real Returns
  12. 12LP Suitability Index
The analytical engine

143 modules. Zero LLM math.

Every node below is one deterministic module in the shipping engine, grouped by theme. AI reads documents; it never computes a number. Select a module to read what it checks, then open the worked example to see the output on one deal.

Macro/Rates

10 modules

Capital Markets

14 modules

Climate/ESG

13 modules

Sector Demand

20 modules

Tax/Cost

9 modules

Operations

45 modules

Portfolio

8 modules

Mandates

1 module

Valuation

9 modules

Risk

11 modules

Capstones

3 modules
Choose a starting point

One desk, three ways in.

Start with the material already on your desk. Every path lands in the same model, scenario set, export, and memo workflow.

01

Upload an IM or pro forma

Bring a PDF, Excel workbook, CSV rent roll, or pasted broker text. Review every extracted field before it reaches the model.

Start from a file
02

Build from assumptions

Start with the underwriting desk, replace the Cayuga Commons assumptions, and watch returns and risk recompute as you type.

Open the inputs
03

Read a completed IC memo

Inspect a full deal before entering your own data. The Cayuga Commons memo shows the output and audit trail end to end.

Review the demo memo
Fast is useful.
Auditable is the product.

Paid beta access starts at $49 per month.

No automatic charge. Ali reviews fit and scope before activation. The Cayuga Commons demo, public methodology, and shipping log remain open for review.

Analyst
$49/mo
Pro
$149/mo
Team
$399/mo
Review pricing

Leave a work email and Ali follows up on fit and scope. No automatic charge.

Built from the desk, not around it.

“The first pass through a CRE deal should not cost a half-day in Excel, and it should not cost the audit trail. Underwrite is the desk I wanted to use myself: fast enough for triage, defensible enough for IC.”

Ali Azim | Founder | Cornell MPS-RE candidate

Try the Cayuga Commons demoSee what shippedEmail Ali directly