top of page
ChatGPT Image Apr 27, 2026, 12_17_04 AM_
ChatGPT Image Apr 27, 2026, 12_19_33 AM.png

OVERVIEW

This is a 3D stealth-based game, where the player takes on the role of a renter trying to gather evidence against their landlord.

 

The house contains several broken appliances that consume excessive energy, but the landlord refuses to fix them, leading the renter to collect proof for a legal case.

Players must explore the house, take photos of faulty appliances, and avoid being detected by the landlord while moving through different rooms. If the player gets caught three times, they lose the game.

Engine 

Unity, C#

ChatGPT Image Apr 30, 2026, 07_07_55 PM.png

Platform 

Windows

Project Type

Team of 5

Primarily Role

Gameplay Programming

School Duration

4 Months

The game begins with a diary that introduces the story and provides instructions to the player.

 

After that, the player starts in their room, where they can see key UI elements:

1. a catalog icon in the bottom-right corner,

2. a strike counter (maximum of three) in the top-right corner, 

3. a timer showing the time of day,

4. and a camera icon on the screen also guides players on how to take photos of the evidence.

be4cafdd-84ca-4a42-a0b4-c426b9d0459b.png

The catalog helps guide the player in two ways:

1. One page shows the landlord’s daily schedule, allowing players to track his movements and avoid certain areas at specific times.

2. Another page provides clues about four pieces of evidence, including hints about their locations and appearance, along with empty slots for the photos that need to be collected.

MY CONTRIBUTIONS

Loading Screen System

  1.  Implemented loading screen UI for scene transitions

Evidence Journal System

  1.  Implemented raycasting system for detecting evidence objects

  2.  Used Unity screenshot functionality to capture evidence on key input (“T”)

  3.  Added notification system to confirm successful captures

  4.  Created journal UI to store and display collected evidence

  5.  Enabled journal access via UI icon and keyboard input (“I”)

Dynamic Suspense AI System

  1.  Implemented landlord AI using state machine and behavior tree logic

  2.  Added waypoint-based patrol and idle behaviors

  3.  Developed detection and chase system when player is spotted

  4.  Integrated AI reactions with gameplay events (evidence capture)

  5.  Created tension through dynamic AI behavior and player interaction

As players explore the house, they must find and photograph the correct evidence. Each successful photo triggers a notification confirming that the evidence has been captured. In addition, there are specific interactive objects placed around the environment that players can use to temporarily freeze the landlord, giving them more time to explore safely and collect evidence.

If the landlord spots the player, he will chase them, giving the player a chance to escape.

 

However, if he catches the player, they receive a strike, and after three strikes, the game is lost.

 

Only correct evidence appears in the catalog, so taking photos of random objects will not count. Once all four pieces of evidence are found and photographed, the game is successfully completed.

bottom of page