This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine
What I Built
What if the things we write down never really disappear, even when we forget where we saved them?
For Week 4's Ghost in the Machine theme, I wanted to explore the connection between technology, memories, and the information we leave behind.
I extended NoteOcean, my open-source AI-powered note-taking application, with a dedicated experience called Ghost Notebook.
The idea is to create a separate, atmospheric notebook space inspired by forgotten memories, old thoughts, unfinished ideas, and stories worth preserving.
The ghost isn't meant to be something supernatural. It's a metaphor for the information that remains hidden inside our digital spaces, waiting to be discovered again.
Ghost Notebook brings this idea into NoteOcean through:
- A dedicated ghost-themed notebook experience.
- A darker, atmospheric visual identity inspired by the Ghost in the Machine theme.
- A familiar rich-text editing experience with Markdown shortcuts and slash commands.
- NoteOcean Bot, which uses open-weight AI to help retrieve and understand information from saved notes.
- Contextual answers with references to the original pages.
The goal is simple: make forgotten information feel alive again.
Demo
Live application: https://noteocean-l222m.ondigitalocean.app/
The Ghost Notebook experience will be available as a separate section inside NoteOcean.
Code
GitHub repository:
NoteOcean 🌊
A modern, fast, AI-powered personal workspace inspired by Notion, featuring instant Markdown transformations, slash commands, real-time title search, browser-local persistence, and an AI assistant strictly grounded in your personal notes.
🌐 Live Application: https://noteocean-l222m.ondigitalocean.app/
📄 Hackathon Presentation: View Slide Deck (PDF)
📸 Screenshots & UI Preview
| NoteOcean Workspace Canvas | Grounded AI Assistant (NoteOcean Bot) |
|---|---|
 |
 |
| Fast, zero-login workspace featuring instant Markdown transformations, editable code blocks, and real-time title search. | NoteOcean Bot answering queries strictly grounded in saved personal notes with clickable source citations. |
📊 Presentation
The official NoteOcean hackathon slide deck details the problem statement, product features, editor interactions, grounded retrieval architecture, privacy boundaries, open-weight Google Gemma integration, and DigitalOcean App Platform deployment.
🌊 Overview & Product Vision
NoteOcean is built for software engineers, students, and writers who want the rapid, keyboard-first velocity of Markdown coupled with rich block layouts…
The project is released under the MIT License.
How I Built It
NoteOcean is built using:
- Next.js and TypeScript for the application.
- Tailwind CSS for the interface.
- Tiptap and ProseMirror for rich-text editing.
- Browser localStorage for note persistence.
- Backboard API for AI integration.
- Google's Gemma-family open-weight model for answering questions.
- DigitalOcean App Platform for deployment.
The interesting part is how the AI works.
When someone asks a question, NoteOcean searches their saved notes, retrieves relevant content, and sends the selected context to a server-side API endpoint.
Backboard routes the request to the configured Gemma model, which generates an answer using the retrieved information.
The response is displayed with references to the relevant notebook pages.
The model is not simply generating random ghost stories. It is helping users rediscover information they have already written down.
Notes are stored locally in the browser. Relevant excerpts are sent to the configured AI provider when a user asks a question. The application does not currently provide cross-device synchronization or offline AI inference.
Why Does Open Innovation Matter?
I believe open-weight AI makes it possible for developers to experiment beyond conventional chatbot experiences.
With models such as Gemma, developers can explore how AI interacts with personal knowledge, build specialized workflows, and change or replace models as their requirements evolve.
For NoteOcean, open-weight AI is an important part of the actual experience. Without it, the application would simply be a notebook with search. With it, users can ask questions in natural language and retrieve contextual answers from their own information.
Backboard provides the API integration, while the application controls retrieval, context selection, and source validation.
This approach also keeps the architecture relatively small and gives the project flexibility to evolve.
The Spirit of It
We write down hundreds of little things throughout our lives.
An idea during a late-night coding session. A conversation we don't want to forget. An address, a thought, a plan, or a story.
Eventually, we forget where we put them.
But the information is still there.
I liked the idea of interpreting a ghost as something from the past that hasn't truly disappeared. A thought that has been sitting quietly in a notebook, waiting for someone to bring it back.
That's what Ghost Notebook represents to me.
A little space where forgotten thoughts can find their way back into the present.
Prize Categories
- Best Use of DigitalOcean: NoteOcean is deployed using DigitalOcean App Platform.
- Best Use of Gemma: Using a Gemma-family open-weight model for contextual question answering.
- Best Use of Backboard: Using Backboard's API to power the AI assistant.
Closing Thoughts
The Ghost in the Machine theme gave me an opportunity to explore something beyond another conventional productivity feature.
I wanted to build an experience that connects memories, personal knowledge, and open-weight AI.
Sometimes, the most interesting things inside a machine are the things we thought we had forgotten.
NoteOcean — Make your notes think with you.
Top comments (0)