Case Study 03 · KBP Investments LLC · Interactive Systems Simulation

Control

I turned real-estate capital allocation into a stateful strategy simulation where decisions compound: acquire assets, choose protection, manage leverage, improve condition, react to market shocks, negotiate with AI opponents, preserve liquidity, and convert holdings into paid-off cash-flowing “Dragons.”

Launch Recruiter Demo ↗Back to Portfolio
Client / Product SponsorKBP Investments LLCAffiliated company · product / learning initiative
Product TypeInteractive financial / strategy simulation
My RoleProduct strategy, UX architecture, interaction logic, AI-assisted prototyping
Core ProblemMake second-order financial consequences visible through play
Public BuildSolo recruiter demo · 1 human vs 3 AI · 1d20
The Problem

Static advice does not expose the operating model.

Capital allocation is a sequence problem. A purchase changes cash. Financing changes liquidity and future options. Protection choices consume cash but alter modeled risk. Renovation changes asset condition, rent, and resale value. External events expose weak reserves, unmanaged properties, or excessive leverage.

The design problem was to turn those dependencies into a system people could experience rather than a list of concepts they had to memorize.

What this proves

  • State-machine and rules-based product thinking
  • Complex cause-and-effect UX across financial states
  • Progressive disclosure across dense management workflows
  • Behavior-driven AI opponents rather than static scripted bots
  • Learning design through consequences, feedback, and recovery
The interface is not the product. The product is the network of rules underneath it: cash, assets, debt, risk, condition, market state, opponent behavior, and the next decision available to the player.
Core Loop

Six decisions reinforce the same strategic model.

01AcquireEvaluate opportunities and decide when to deploy scarce cash.
02ProtectModel tradeoffs around LLC, insurance, and property-management choices.
03ImproveMove rentals from Distressed → Fair → Good → Turnkey.
04FinanceMortgage or pay off assets, trading liquidity against income and risk.
05AdaptRespond to market events, losses, auctions, trades, and cash pressure.
06CompoundTurn paid-off income assets into Dragons and progress toward King Dragon.
Systems Architecture

What the simulation actually models

01 · ASSET STATE

Condition changes economics

Rental upgrades move through explicit condition tiers. Improvement spending raises modeled rent and resale value, making forced appreciation visible as a state transition.

02 · CAPITAL

Liquidity has opportunity cost

Cash, mortgage proceeds, payoff amounts, development spending, recurring fees, rent, and equity interact. The player must decide what to hold, deploy, protect, or de-lever.

03 · RISK

Protection is a decision, not decoration

The game models entity structure, insurance, and property management as spend-versus-resilience choices that affect later event outcomes.

04 · AI BEHAVIOR

Opponents have operating doctrines

AI profiles vary buying aggression, willingness to mortgage, auction ceilings, reserve targets, build appetite, trade behavior, and protection choices.

05 · MARKET STATE

External conditions change the board

Recurring market events can alter rent, payoff pressure, vacancy exposure, property value assumptions, and other conditions across all active players.

06 · FAILURE / RECOVERY

Cash pressure creates meaningful branches

When obligations exceed cash, players can mortgage assets, liquidate improvements, pay obligations, or declare bankruptcy. State must remain coherent through those transitions.

AI Opponents

Different strategies create different system pressure.

The Shark

High acquisition aggression, heavy leverage tolerance, minimal cash buffer, aggressive development, and little modeled protection. It creates upside pressure and fragility.

The Vault

Conservative purchasing, high reserve threshold, low mortgage appetite, insurance-first protection, and stricter auction discipline. It creates a defensive benchmark.

Portfolio Hunters

Other personalities target strategic groups, trade to complete positions, and alter development or protection decisions using personality-specific parameters.

Decision Surfaces

Complexity is distributed instead of dumped onto one screen.

Management layer

  • Cash and protection-spend tracker
  • Property management and asset controls
  • Mortgage management
  • Renovation / development
  • Portfolio and progression view

Exception layer

  • Auctions and trades
  • Market-event consequences
  • Payment pressure and bankruptcy recovery
  • AI trade proposals
  • Acquisition SITREP and decision modals
1 + 3Human + AI demo configuration
8Parameterized AI personality roster in the full engine
4Rental condition states
25Dragons for King Dragon ledger status
Recruiter Walkthrough

A five-minute route through the strongest evidence

01 · Start

Launch Solo Demo. The public build is intentionally locked to one human, three AI opponents, and d20 movement to remove setup noise.

02 · Acquire

Land on an opportunity and inspect the purchase / structure decision. Notice that the acquisition changes downstream options.

03 · Manage

Open property management, mortgage, renovation, or portfolio views. The point is linked state, not isolated screens.

04 · Observe AI

Watch opponents buy, hold reserves, finance, build, auction, or trade according to different behavior profiles.

05 · Absorb shock

Market and risk events expose prior choices. Weak liquidity or protection creates different consequences from stronger structure.

06 · Read the model

Return to the portfolio / cash surfaces and see how prior actions changed liquidity, equity, income assets, and progression.

Simulation boundary

Control intentionally simplifies real-estate, legal, tax, financing, insurance, and property-management concepts into visible game mechanics. The portfolio value is the interaction model and consequence architecture—not factual investment or legal guidance.

Interactive Evidence

Run the system

The embedded build is the same public recruiter demo linked above. It uses isolated local browser storage and a simplified Solo + d20 setup.

Control · recruiter demoSolo vs 3 AI · 1d20 · illustrative simulation
Transferable UX Value

This is systems thinking in a different surface.

Control sits next to Portfolio Command and Property Acquisition Intelligence because all three products are solving the same class of problem: make hidden rules visible, model state changes, expose consequences, and help a person decide what to do next.

Portfolio signal

  • Product architecture across many dependent states
  • Information design for high-choice environments
  • Rules, exceptions, recovery, and edge-case handling
  • Behavioral simulation and AI-agent parameter design
  • Front-end-aware prototyping at substantial product depth