Yanah AI

Yanah AI turns written stories into playable 3D worlds, automatically generating environments, music, characters, and gameplay so anyone can step inside the worlds they imagine.

Comment 4

Inspiration

The Problem: Millions of people write stories every day, but almost none of those stories ever become worlds that can be explored. Turning imagination into an interactive experience requires learning complex tools like game engines, 3D modeling, and level design.

Our Solution: We use AI to remove that barrier entirely.

Yanah AI does not just generate stories. It transforms them into fully playable worlds. A simple prompt becomes an environment that users can walk through, explore, and experience.

Storytelling today is passive. You write and you read, and that is where it ends.

We wanted to change that.

What if a story did not stay on the page What if it became a world you could step inside

That question became Yanah AI. A system where imagination instantly turns into exploration.

The most rewarding moment during development was watching a simple paragraph transform into a living, explorable world with characters, environments, and narrative unfolding in real time.

Technology is often seen as complex and inaccessible in game creation.

But what if it could make creation effortless

Yanah AI is built on a simple belief. With the right architecture, AI can turn anyone into a world builder instantly.

What it does

Yanah AI is a Story to World Engine built around Gemini’s ability to understand, expand, and structure narrative into spatial experiences.

A user writes a simple story prompt.

From that input, the system:

Expands the story using Gemini including characters, lore, and narrative structure Converts the story into a structured world graph Translates that graph into a spatial layout with coordinates Generates visual assets, environments, and atmosphere Assembles everything into a playable browser based world

The result is a complete interactive environment generated from a single idea.

Key capabilities:

Instant world generation from story to playable environment Dynamic NPCs powered by Gemini with goals, emotions, and context awareness An explorable narrative where players uncover the story through movement and interaction Replayability, where every new prompt creates a new world No game engine required, everything runs directly in the browser

Instead of reading a story, users experience it.

How we built it

This project was built as a fully automated AI pipeline on Google Cloud Platform, designed to transform unstructured text into structured, playable worlds.

Architecture Overview

Backend Pipeline on Cloud Run

Service 1 receives the story input

Service 2 sends the prompt to Gemini on Vertex AI

Service 3 builds a structured world graph

Service 4 converts the graph into a spatial layout with coordinates

Service 5 generates assets including visuals, narration, and music

AI Models

Gemini for narrative expansion and reasoning Image generation for environments and objects LLM powered agents for NPC dialogue

Storage and Deployment

Cloud Storage stores assets and world data Firebase Hosting deploys the playable world to the browser Core Innovation

The system connects three layers:

Narrative which is the story Structure which is the world graph Space which is the playable environment

This transformation pipeline makes instant world generation possible.

Challenges we ran into

Turning stories into structured systems Narratives are ambiguous. Converting them into consistent world graphs required careful prompt design and validation logic.

Maintaining coherence across space Ensuring that locations, characters, and events made sense spatially as well as narratively.

Real time generation constraints Balancing speed and quality while keeping the experience responsive.

NPC realism Making characters feel alive required designing them with goals, memory, and context awareness.

Pipeline orchestration Coordinating multiple services into a seamless flow from story to deployment.

Accomplishments that we are proud of

A complete story to a playable world pipeline Real-time browser-based world deployment Dynamic NPCs that behave like real characters A scalable system built on Google Cloud A platform where anyone can create a world without technical skills

What we learned

Stories alone are not enough; structure is critical Transforming imagination into interaction requires systems thinking

AI works best as a pipeline Breaking the system into stages improved reliability and quality

Exploration changes engagement Interacting with a story creates a deeper experience than reading it

Simplicity is powerful Typing a story and getting a world is the core value

What is next for Yanah AI

Phase 1 Creator Tools World editing tools Save and share functionality Persistent story progression

Phase 2 Social Worlds Multiplayer exploration Collaborative storytelling Shared world building

Phase 3 Advanced AI Systems Improved NPC memory and behavior Adaptive storylines based on player actions Personalized world generation

Phase 4 Platform Expansion Mobile support VR and AR integration Creator marketplace

The Bigger Vision

If anyone can turn imagination into a world instantly, we unlock a new form of creativity.

A future where Writers become world builders Stories become experiences Imagination becomes something you can explore

Yanah AI is not just a tool

It is a new way to experience ideas

The goal

Millions of people stepping inside their own stories

Built With

  • cloud-run-microservices
  • cloud-storage
  • firebase-hosting
  • gemini-2.0-flash
  • gemini-tts
  • gemini-vision
  • godot-engine
  • google-cloud
  • graph
  • imagen-3
  • json
  • lyria-music-generation
  • python
  • rest-apis
  • vertex-ai
  • world
Share this project:

Updates