HARTZAI
Book Discovery Call
Adversarial agents

Watch seven AI experts pressure-test a deal, then agree on what to do

Most AI on the web is a chatbot answering a question. This is a bench of specialist agents pressure-testing the same deal and reaching a call they can defend, then drafting the document an investor would expect. Press one button and watch it happen.

Commercial real estateIllustrative data

Multi-let office, regional city centre

An off-market acquisition put to the investment committee.

The decision on the table

Buy at the asking price, chip the price, or walk away?


Deal data room
Asset
Location
Regional city centre, UK
Asset type
Multi-let office
Tenure
Freehold
Net internal area
38,500 sq ft
Suites
14 (11 let, 3 vacant)
Car parking
32 spaces
EPC
Average D, two suites E
Pricing
Asking price
£8,200,000
Recommended offer
£7,540,000
Capital value
£196 per sq ft
Net initial yield
7.4% asking / 8.0% offer
Reversionary yield
9.3%
Income
Passing rent
£607,000 pa
ERV, fully let
£739,000 pa
WAULT
4.2 years to expiry
Occupancy
86% by floor area
Largest tenant
Northgate Advisory, 37% of rent
Debt & returns
Debt offered
60% LTV at 6.1%
Day-one ICR
2.2x
Target net IRR
Circa 12% over 5 years
Equity multiple
Circa 1.7x
The model
Acquisition_Model_v3.xlsxRead-only
A1fxIllustrative model
 AB
1MetricValue
2Recommended price£7,540,000
3Acquisition costs (6.8%)£513,000
4Total outlay£8,053,000
5Senior debt (60% LTV)£4,524,000
6Equity required£3,529,000
7Day-one NOI£607,000
8Day-one ICR2.20x
9Day-one cash-on-cash9.4%
10Stabilised NOI£739,000
11Exit value (year 5)£9,700,000
12Net IRR (5-year)12.0%
13Equity multiple1.70x

Illustrative model. Exit modelled at a 7.5% net initial yield on stabilised income.

Acquisition Panel
7 membersGroup chat

Press Run the debate to start the conversation.

Seven specialists join the chat. Watch them pressure-test the deal in real time.

Illustrative and scripted. The agents do not call a live model here, the figures and spreadsheets are invented, and this describes no real deal or company. In a real build the agents run on your own data, with a person making the final decision.

How it works

Seven experts, two rounds, one defensible call

This is not a chatbot answering a question. It is seven AI agents, each playing a property specialist with its own remit, pressure-testing the same deal until they reach a conclusion they can all stand behind, then turning it into a board-ready document.

  1. Step 01

    Seven specialists, seven angles

    A valuer tests the price, a building surveyor prices the works, a solicitor reads the leases, an accountant structures for tax, a broker sources the debt, an environmental consultant weighs the risks, and a planning consultant maps what you can do.

  2. Step 02

    Two rounds of argument

    In the first round each specialist gives their read. In the second the tensions surface and resolve, which is where the real conditions appear.

  3. Step 03

    A verdict, then the paperwork

    A final pass weighs the exchange into a recommendation with conditions and options, then drafts the full Information Memorandum behind it.

No overclaiming

In a real build, these agents are grounded in your own data, not invented figures, and they draft the decision and the document rather than make the call. A person always signs it off. This page is scripted to show the shape of that reasoning, and every figure on it is illustrative.

The honest bit

This pattern applies to your underwriting, your hiring shortlists, your supplier decisions, any call where being able to defend the answer matters as much as the answer itself.

See what this could do for your decisions

Before you ask

The honest questions about a demo like this

Where to next

See the pattern at work