This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine
What I Built
GhostMemory AI is a spooky, AI-powered storytelling experience inspired by the idea that memories never truly disappear—they can return as stories, fragments, and echoes from the past.
For Week 4's Ghost in the Machine theme, I wanted to explore the boundary between memory, imagination, and artificial intelligence. What if an old memory could become a mysterious story? What if an unfinished diary entry could be reimagined into something haunting?
GhostMemory AI is designed around that idea, with a dark, atmospheric interface and features for creating eerie stories, keeping memories, reimagining diary entries, and exploring saved stories.
Key Features
👻 Create Story: Turn a premise or idea into a paranormal mystery.
🕯️ Memory Keeper: Explore the concept of preserving meaningful memories.
📖 Diary Reimaginer: Transform diary entries and forgotten moments into new narratives.
🌌 My Stories: Keep storytelling experiences together.
🤖 Cloud AI Integration: Connect the application to Cloudflare Workers AI through server-side functions.
Demo
Live Website: https://ghostmemory-ai.netlify.app/
Code
GitHub Repository: https://github.com/priyarani22023-max/GhostMemory-AI
How I Built It
I developed GhostMemory AI using Google AI Studio to help build and refine the application, then connected the code to GitHub and deployed the website through Netlify.
My planned AI architecture uses Cloudflare Workers AI for hosted inference with an open-weight model. Netlify Functions handle server-side requests so that API credentials remain outside the frontend.
The intended flow is:
A user enters a story premise or memory.
The frontend sends a request to a Netlify Function.
The function calls the configured Cloudflare Workers AI model.
The generated result is returned to the frontend and displayed to the user.
The application is being configured to use environment variables for the Cloudflare Account ID, API token, and model ID rather than hardcoding secrets in the source code.
I chose a serverless architecture so that the project can be accessed online without requiring visitors to install a local inference engine.
Why Does Open Innovation Matter?
Open innovation makes experimentation with AI more accessible. Developers can explore open-weight models, build their own interfaces, and choose how inference is hosted instead of relying entirely on a single closed API.
For GhostMemory AI, this approach creates an opportunity to experiment with creative storytelling while learning about model integration, serverless functions, deployment, and secure credential management.
It also makes the project easier to inspect, improve, and extend with community contributions.
Top comments (0)