The habit of keeping a diary is rapidly disappearing. Leather-bound diaries and composition books once served as repositories for everything from mundane daily observations to the deepest existential crises. To record these thoughts and observations, a person needed to sit quietly, reflect on his experiences and write his inner world on paper.
Today, much of the lives we examine has moved from personal pages to public social media feeds. With this change, writing became performative rather than speculative. The deliberate pace of handwriting has also been accelerated by the fast pace of digital tools. The bottom line is that the habit of self-reflection is fading, and that’s not good for any of us.
Hardware (📷: Crystal Orion)
GitHub user cristal-orion still sees journaling as an important habit, and has developed a distraction-free digital tool to help you get the most out of it. Called Reminor, it’s an AI-powered personal diary. You write down your thoughts, and Reminor analyzes them, recalls them, and chats with you about them. These are all things that can help you get more information out of diary entries that can get lost among the hundreds of pages that accumulate over the years.
Unlike most modern AI journaling platforms, Reminor is not a website or subscription service. Designed to run locally on the Raspberry Pi. The system hosts its own web interface and backend and stores all entries directly on your device. Users type on an external keyboard, and cristal-orion uses a 4-inch Waveshare LCD mounted inside a 3D-printed case to create a purpose-built writing instrument that sits quietly on a desk and consumes less than 5 watts of power.
Under the hood, the Pi runs a Docker-based software stack consisting of a Svelte frontend and a FastAPI backend. Journal entries are saved as simple dated text files. Reminor processes each entry using a local embedding model that converts writes into mathematical vectors (numeric representations of meaning). These vectors are stored using a system called Memvid. Memvid compresses vectors into video frames to save storage while allowing semantic search. Instead of searching for specific words, users can ask questions like, “When was the last time you started worrying about your job?” Then search for related passages written months or years apart.
Enlarged view of the device (📷: cristal-orion)
The system also tracks emotional tone across eight dimensions, visualizing long-term mood trends and streaks. An integrated chat interface allows users to ask questions about their past. The software retrieves relevant entries and uses AI models to summarize patterns and connections. Importantly, all of this can be done offline using local models. Remote large-scale language model APIs from providers such as OpenAI, Anthropic, and Google are optional and will only be used if you explicitly enable them.
If you would like to try Reminor for yourself, first GitHub repository.—
