This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine
What I Built
Meet memly — Idea & Memory Sanctuary, an open-source space designed to help you capture, preserve, revisit, and rediscover your thoughts and memories.
We often collect ideas, notes, reflections, and little moments throughout our lives. But many of them get buried in folders, forgotten apps, or scattered documents.
I wanted to build something different: a personal sanctuary where memories aren't just stored — they can be revisited and explored.
✨ Key Features
📝 Memory Capture: Save thoughts, ideas, reflections, and personal moments.
🧠 AI Memory Synthesis (Echoes): Explore connections and recurring themes in your saved memories.
🪐 Memory Constellation: Discover relationships between memories through a visual experience.
🕰️ Timeline: Revisit your memories over time.
🎙️ Voice Capture: Capture thoughts using voice.
💬 AI Reflections: Ask questions and explore your archive using AI.
📦 Export Options: Preserve your archive in JSON, Markdown, and CSV formats.
Demo
1.main screen
2.local ai connection
Code
tarun-sane
/
memly
👻 memly — Idea & Memory Sanctuary | Give Your Forgotten Memories a Second Life with Open-Source AI
Meet memly — Idea & Memory Sanctuary, an open-source space designed to help you capture, preserve, revisit, and rediscover your thoughts and memories.
We often collect ideas, notes, reflections, and little moments throughout our lives. But many of them get buried in folders, forgotten apps, or scattered documents.
I wanted to build something different: a personal sanctuary where memories aren't just stored — they can be revisited and explored. ✨ Key Features
- 📝 Memory Capture: Save thoughts, ideas, reflections, and personal moments.
- 🧠 AI Memory Synthesis (Echoes): Explore connections and recurring themes in your saved memories.
- 🪐 Memory Constellation: Discover relationships between memories through a visual experience.
- 🕰️ Timeline: Revisit your memories over time.
- 🎙️ Voice Capture: Capture thoughts using voice.
- 💬 AI Reflections: Ask questions and explore your archive using AI.
- 📦 Export Options: Preserve your archive in JSON, Markdown, and CSV formats.
How I Built It
🛠️ How I Built It
The project uses:
- React 19
- TypeScript
- Vite
- Tailwind CSS
- Node.js and Express
- Browser LocalStorage for memory persistence
- Ollama / compatible local AI endpoints for AI-powered experiences
- Optional Gemini integration I wanted to explore how open-source AI could make personal knowledge management more meaningful while keeping control of personal data in the user's hands.
🤖 Local AI Integration
One of the interesting parts of memly is its local AI direction.
Instead of making cloud AI the only option, the project supports connecting to local inference services such as Ollama and compatible OpenAI-style endpoints.
This allows users to experiment with models running on their own machines.
🚀 How to Run the Project
Clone the repository:
git clone YOUR_GITHUB_REPOSITORY_URL
cd memly
Install dependencies:
npm install
Start the development server:
npm run dev
copy the url and open it in the web browser .
If you're using local AI through Ollama, install Ollama, pull a compatible model, and configure the AI endpoint according to the project's setup instructions.
Check the README for the exact model and configuration details.
Why Does Open Innovation Matter?
I believe personal knowledge tools should be accessible, adaptable, and transparent.
Open-source development allows people to inspect, improve, customize, and build upon ideas together.
With memly, I wanted to explore a simple question:
What if our forgotten thoughts could become a source of new discoveries?
The Spirit of It
memly is more than a place to save notes. It's an experiment in remembering, reflecting, and finding meaning in the things we collect along the way.
I'd love to hear your thoughts, suggestions, and ideas for improving it!
If you find the project interesting, please consider giving it a ⭐ on GitHub.
Thanks for reading! 💚
Prize Categories
Best Use of Gemma -- memly explores local AI-powered memory discovery and reflection using open-weight models through Ollama.
Top comments (0)